Publish complete SnapTop model catalog

Publish verified speech recognition, alignment, OCR, neural denoising and development translation models alongside the existing SAM2, SAM3 and narration payloads. Include original licenses, upstream provenance, file checksums and a complete catalog.

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+madlad-translation/tokenizer.json filter=lfs diff=lfs merge=lfs -text

LICENSES.md

@@ -0,0 +1,19 @@
+# Model-specific licenses
+
+SAM3 root weights and ONNX archive: [Meta SAM License](LICENSE).
+
+| Model | License | Original source |
+| --- | --- | --- |
+| [chatterbox-turbo](chatterbox-turbo) | [mit](chatterbox-turbo/LICENSE) | [ResembleAI/chatterbox-turbo](https://huggingface.co/ResembleAI/chatterbox-turbo/tree/749d1c1a46eb10492095d68fbcf55691ccf137cd) |
+| [madlad-translation](madlad-translation) | [apache-2.0](madlad-translation/LICENSE) | [google/madlad400-3b-mt](https://huggingface.co/google/madlad400-3b-mt/tree/fa184c675da0b5c9e1c8694fccd4e12e2d422094) |
+| [qwen-aligner](qwen-aligner) | [apache-2.0](qwen-aligner/LICENSE) | [Qwen/Qwen3-ForcedAligner-0.6B](https://huggingface.co/Qwen/Qwen3-ForcedAligner-0.6B/tree/c7cbfc2048c462b0d63a45797104fc9db3ad62b7) |
+| [qwen-asr](qwen-asr) | [apache-2.0](qwen-asr/LICENSE) | [Qwen/Qwen3-ASR-1.7B](https://huggingface.co/Qwen/Qwen3-ASR-1.7B/tree/7278e1e70fe206f11671096ffdd38061171dd6e5) |
+| [rnnoise](rnnoise) | [bsd-3-clause](rnnoise/LICENSE) | [https://raw.githubusercontent.com/richardpl/arnndn-models/0fda24c46d78f0207d820bb970fbe85c1971b39c/std.rnnn](https://raw.githubusercontent.com/richardpl/arnndn-models/0fda24c46d78f0207d820bb970fbe85c1971b39c/std.rnnn) |
+| [sam2-base-plus](sam2-base-plus) | [apache-2.0](sam2-base-plus/LICENSE) | [facebook/sam2.1-hiera-base-plus](https://huggingface.co/facebook/sam2.1-hiera-base-plus/tree/b7320756a13354e7530a63935656d35b2f91a290) |
+| [sam2-large](sam2-large) | [apache-2.0](sam2-large/LICENSE) | [facebook/sam2.1-hiera-large](https://huggingface.co/facebook/sam2.1-hiera-large/tree/665f8e2ad61cf5f53d65644ff27c8ee525124610) |
+| [sam2-small](sam2-small) | [apache-2.0](sam2-small/LICENSE) | [facebook/sam2.1-hiera-small](https://huggingface.co/facebook/sam2.1-hiera-small/tree/ee5bba1d82bb8749febdf90f45e84b687142ba03) |
+| [sam2-tiny](sam2-tiny) | [apache-2.0](sam2-tiny/LICENSE) | [facebook/sam2.1-hiera-tiny](https://huggingface.co/facebook/sam2.1-hiera-tiny/tree/de431c4043854a71d8101e17995dfe596bf101a5) |
+| [tesseract-eng](tesseract-eng) | [apache-2.0](tesseract-eng/LICENSE) | [https://cdn.jsdelivr.net/npm/](https://huggingface.co/https://cdn.jsdelivr.net/npm//tree/tesseract.js-data/eng/4.0.0_best_int/eng.traineddata.gz) |
+| [unlimited-ocr](unlimited-ocr) | [mit](unlimited-ocr/LICENSE) | [baidu/Unlimited-OCR](https://huggingface.co/baidu/Unlimited-OCR/tree/8249d20b2822a7d8a4c22c8d8c9069b5ac36bda3) |
+| [whisper-large-v3](whisper-large-v3) | [mit](whisper-large-v3/LICENSE) | [Systran/faster-whisper-large-v3](https://huggingface.co/Systran/faster-whisper-large-v3/tree/edaa852ec7e145841d8ffdb056a99866b5f0a478) |
+| [whisper-small](whisper-small) | [mit](whisper-small/LICENSE) | [Systran/faster-whisper-small](https://huggingface.co/Systran/faster-whisper-small/tree/536b0662742c02347bc0e980a01041f333bce120) |

MODEL_CATALOG.json

@@ -0,0 +1,529 @@
+{
+  "repository": "hacksider/snaptop-models",
+  "sam3": {
+    "python": {
+      "LICENSE": {
+        "size": 7352,
+        "sha256": "bec48f70bd37bf8280a9d1ebf01642d26086f6122ba735baa08fe03c5a6e7448"
+      },
+      "config.json": {
+        "size": 25843,
+        "sha256": "4616385e4b21f2e5e22c875b65679185cbccfa95de42542b9166f7dc3d57160f"
+      },
+      "merges.txt": {
+        "size": 524619,
+        "sha256": "9fd691f7c8039210e0fced15865466c65820d09b63988b0174bfe25de299051a"
+      },
+      "model.safetensors": {
+        "size": 3439938512,
+        "sha256": "6d06f0a5f84e435071fe6603e61d0b4cc7b40e0d39d487cfd4d67d8cc11cc14a"
+      },
+      "processor_config.json": {
+        "size": 1712,
+        "sha256": "6420cf2671fa9309ea95bc0144a8b9861666d1c5f43c8db09e410dacda974fce"
+      },
+      "special_tokens_map.json": {
+        "size": 588,
+        "sha256": "2cdb3b8331a60c92fc1e55a13e9fd61fd2293c5a51275fdcccd62b780052530e"
+      },
+      "tokenizer.json": {
+        "size": 3642073,
+        "sha256": "6d9109cc838977f3ca94a379eec36aecc7c807e1785cd729660ca2fc0171fb35"
+      },
+      "tokenizer_config.json": {
+        "size": 799,
+        "sha256": "39670ad98457fe8f14ca59f6bb74591e9fc850974380a63993e5b8ffc865baa2"
+      },
+      "vocab.json": {
+        "size": 862328,
+        "sha256": "5047b556ce86ccaf6aa22b3ffccfc52d391ea4accdab9c2f2407da5b742d4363"
+      }
+    },
+    "onnx": {
+      "filename": "sam3_vit_h.zip",
+      "size": 3413511219,
+      "sha256": "1617436f9db7eb9dd69029c2c465d54b316c36c21143f0fde0d1be513f15b7d5"
+    }
+  },
+  "catalog": {
+    "chatterbox-turbo": {
+      "prefix": "chatterbox-turbo",
+      "upstream": "ResembleAI/chatterbox-turbo@749d1c1a46eb10492095d68fbcf55691ccf137cd",
+      "license": "mit",
+      "files": {
+        "README.md": {
+          "size": 9544,
+          "sha256": "d007d8c2ed2093530d95eee9d623f8977c321f6f361c19dafa07c000375808d7"
+        },
+        "added_tokens.json": {
+          "size": 418,
+          "sha256": "72e4ab6acb0d9309ac3df4b526ae5fd80a2da5bc5ab7bb02d85096a374f69193"
+        },
+        "merges.txt": {
+          "size": 456318,
+          "sha256": "1ce1664773c50f3e0cc8842619a93edc4624525b728b188a9e0be33b7726adc5"
+        },
+        "special_tokens_map.json": {
+          "size": 470,
+          "sha256": "92ba8063bf40aa163eadebbfe0de07c2aebe44cf0d4a9e8726580b0781fd2640"
+        },
+        "tokenizer_config.json": {
+          "size": 3878,
+          "sha256": "bca16a2ac1ddbd78b8d6228f0031884cc74b6ea54b967d6f6d2ebae9ccde23e6"
+        },
+        "vocab.json": {
+          "size": 999186,
+          "sha256": "f6bd25a65e4e63ca31360e9fb11c7e4f9a391a78385d640acd814092dd6eee4f"
+        },
+        "conds.pt": {
+          "size": 169454,
+          "sha256": "b1852099306fd6a7814eb9d0bd10186caba7249596cc23868f78a0eefbfa5033"
+        },
+        "ve.safetensors": {
+          "size": 5695784,
+          "sha256": "f0921cab452fa278bc25cd23ffd59d36f816d7dc5181dd1bef9751a7fb61f63c"
+        },
+        "s3gen_meanflow.safetensors": {
+          "size": 1064875036,
+          "sha256": "d65cb687a2ed581ee6cc297e919ffefa63386944f42364ae13b78a594945514f"
+        },
+        "t3_turbo_v1.safetensors": {
+          "size": 1915480052,
+          "sha256": "fcf1f8c1d651bb7e3acd69ee5be269b4ac10c02980b7708213d598bc9f7cdf87"
+        },
+        "LICENSE": {
+          "size": 1067,
+          "sha256": "4248e910a928849fe5815a0f9236e17fa07768d95b9193212752c464b93d6caa"
+        },
+        "PROVENANCE.txt": {
+          "size": 184,
+          "sha256": "633e03dbd0393ca8fb3ee2eb9a1c5ec6708398380d6e713b7e1a64e71b34e327"
+        }
+      }
+    },
+    "madlad-translation": {
+      "prefix": "madlad-translation",
+      "upstream": "google/madlad400-3b-mt@fa184c675da0b5c9e1c8694fccd4e12e2d422094",
+      "license": "apache-2.0",
+      "files": {
+        "config.json": {
+          "size": 749,
+          "sha256": "cad399cab799b99409a6ec2d90d72552257c2bb752861261d2016691e0643e7c"
+        },
+        "generation_config.json": {
+          "size": 142,
+          "sha256": "0849b38987568ccfe4ebefc22bbda1cec4bee01c345e93bfab207d4692b0a1d5"
+        },
+        "added_tokens.json": {
+          "size": 4,
+          "sha256": "d914176fd50bd7f565700006a31aa97b79d3ad17cee20c8e5ff2061d5cb74817"
+        },
+        "special_tokens_map.json": {
+          "size": 414,
+          "sha256": "7f79f1d5063d56c4b980eec0692f3c7429bdef335071d34e566bd00fd4b5e3e0"
+        },
+        "spiece.model": {
+          "size": 4427844,
+          "sha256": "ef11ac9a22c7503492f56d48dce53be20e339b63605983e9f27d2cd0e0f3922c"
+        },
+        "tokenizer.json": {
+          "size": 16629031,
+          "sha256": "a2799ccc696b752ba00c34f58726bfe253a04921ceb6cfc620400f560474790b"
+        },
+        "tokenizer_config.json": {
+          "size": 830,
+          "sha256": "641fc660745306dfb935f666a68f8bc10a44c39241cfb357be518fda8c09662d"
+        },
+        "model.safetensors": {
+          "size": 11761587872,
+          "sha256": "66ff5f8fcaf92291da486fdfbd4d5233cec90e1359348a56e3172c978b3a76d4"
+        },
+        "LICENSE": {
+          "size": 11357,
+          "sha256": "c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4"
+        },
+        "PROVENANCE.txt": {
+          "size": 186,
+          "sha256": "9cfafdae9362a042d581da47691054ea8f5eb8aa628d59ab2866e975c3366f44"
+        }
+      }
+    },
+    "qwen-aligner": {
+      "prefix": "qwen-aligner",
+      "upstream": "Qwen/Qwen3-ForcedAligner-0.6B@c7cbfc2048c462b0d63a45797104fc9db3ad62b7",
+      "license": "apache-2.0",
+      "files": {
+        "README.md": {
+          "size": 57456,
+          "sha256": "5058416891bc47a2051557765997e8c42f8eb78a0e33c3e775bd17d4b0ba4d50"
+        },
+        "chat_template.json": {
+          "size": 1161,
+          "sha256": "75a8cfca24f00de72d796fbfed6858fc9614ef3dabd8696684cc3bc03a9c58ff"
+        },
+        "config.json": {
+          "size": 5982,
+          "sha256": "d616c65d46c4b90bdc651b0a0963ea932732241140f337f9bb6b0335a9c8ef09"
+        },
+        "generation_config.json": {
+          "size": 115,
+          "sha256": "948d089b23bca1d214e768d59c4438365665f52ec6d33678f4062206b3fbbb8c"
+        },
+        "merges.txt": {
+          "size": 1671853,
+          "sha256": "8831e4f1a044471340f7c0a83d7bd71306a5b867e95fd870f74d0c5308a904d5"
+        },
+        "model.safetensors": {
+          "size": 1835544544,
+          "sha256": "47831d0e82f96b20e9034dba01a075ee06436654719f6a68289e49f1b65ce0e7"
+        },
+        "preprocessor_config.json": {
+          "size": 330,
+          "sha256": "45e120a4eda2c20c5d7f2ea9354e63536bf35e27aa573fb7cdf78017b378770d"
+        },
+        "tokenizer_config.json": {
+          "size": 12666,
+          "sha256": "3ab80063f8511deb9566e6ad438d17b7a6277fcffd52d92854112f19d36bd81c"
+        },
+        "vocab.json": {
+          "size": 2776833,
+          "sha256": "ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910"
+        },
+        "LICENSE": {
+          "size": 11357,
+          "sha256": "c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4"
+        },
+        "PROVENANCE.txt": {
+          "size": 193,
+          "sha256": "e536da59d661aeee5e2e1f63825af871f0fb510c2272f213597d5d1945b2c154"
+        }
+      }
+    },
+    "qwen-asr": {
+      "prefix": "qwen-asr",
+      "upstream": "Qwen/Qwen3-ASR-1.7B@7278e1e70fe206f11671096ffdd38061171dd6e5",
+      "license": "apache-2.0",
+      "files": {
+        "README.md": {
+          "size": 57456,
+          "sha256": "5058416891bc47a2051557765997e8c42f8eb78a0e33c3e775bd17d4b0ba4d50"
+        },
+        "chat_template.json": {
+          "size": 1161,
+          "sha256": "75a8cfca24f00de72d796fbfed6858fc9614ef3dabd8696684cc3bc03a9c58ff"
+        },
+        "config.json": {
+          "size": 6194,
+          "sha256": "2e74a751548b8ad7d7526d29365ad8144c345d8b412b1152d25dc6698452712f"
+        },
+        "generation_config.json": {
+          "size": 142,
+          "sha256": "1da527824d81e07118facff437e03f2e24a23311e3bdeb2368973fe77e5f275c"
+        },
+        "merges.txt": {
+          "size": 1671853,
+          "sha256": "8831e4f1a044471340f7c0a83d7bd71306a5b867e95fd870f74d0c5308a904d5"
+        },
+        "model-00001-of-00002.safetensors": {
+          "size": 4220320824,
+          "sha256": "a4cd1f1a04d90b757dc7f7dd26254e69a013b19e80efe590a83c6a3bde8608d6"
+        },
+        "model-00002-of-00002.safetensors": {
+          "size": 478200688,
+          "sha256": "6e0b9d9e09e2e0238e7ef3cc8a484ab387e91b90f1900bedf88bc92d7929ccfc"
+        },
+        "model.safetensors.index.json": {
+          "size": 64821,
+          "sha256": "f994739fe38e5210b9e3e8ce6c6307315e2ceac3cb630e7b7414d69dce520f60"
+        },
+        "preprocessor_config.json": {
+          "size": 330,
+          "sha256": "45e120a4eda2c20c5d7f2ea9354e63536bf35e27aa573fb7cdf78017b378770d"
+        },
+        "tokenizer_config.json": {
+          "size": 12487,
+          "sha256": "4942d005604266809309cabc9f4e9cb89ce855d59b14681fdc0e1cc62ea26c4c"
+        },
+        "vocab.json": {
+          "size": 2776833,
+          "sha256": "ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910"
+        },
+        "LICENSE": {
+          "size": 11357,
+          "sha256": "c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4"
+        },
+        "PROVENANCE.txt": {
+          "size": 183,
+          "sha256": "cfcbdf6e015bf22a5580bbed4e314d8053ad13b880ec1c620700c6aa5f18804d"
+        }
+      }
+    },
+    "rnnoise": {
+      "prefix": "rnnoise",
+      "upstream": "https://raw.githubusercontent.com/richardpl/arnndn-models/0fda24c46d78f0207d820bb970fbe85c1971b39c/std.rnnn",
+      "license": "bsd-3-clause",
+      "files": {
+        "std.rnnn": {
+          "size": 302903,
+          "sha256": "6b8943dc4a9b6b24425873992a44f29c0577503276456af46a8854774faeb294"
+        },
+        "LICENSE": {
+          "size": 1581,
+          "sha256": "e2f59ff41d9d03adc3dcf3deff170f8c8cf4a6eb4a9b174762a7656d23200ffa"
+        },
+        "PROVENANCE.txt": {
+          "size": 148,
+          "sha256": "81933a7bcf5c44b242332e3adafa6140d735c4309e959bb7ea1038156a3ffb6c"
+        }
+      }
+    },
+    "sam2-base-plus": {
+      "prefix": "sam2-base-plus",
+      "upstream": "facebook/sam2.1-hiera-base-plus@b7320756a13354e7530a63935656d35b2f91a290",
+      "license": "apache-2.0",
+      "files": {
+        "sam2.1_hiera_base_plus.pt": {
+          "size": 323606802,
+          "sha256": "a2345aede8715ab1d5d31b4a509fb160c5a4af1970f199d9054ccfb746c004c5"
+        },
+        "LICENSE": {
+          "size": 11357,
+          "sha256": "c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4"
+        },
+        "PROVENANCE.txt": {
+          "size": 195,
+          "sha256": "effcb3a166b276d98dc828e6aede623731ad214ec78fb78dffe8f46ab4b4c979"
+        }
+      }
+    },
+    "sam2-large": {
+      "prefix": "sam2-large",
+      "upstream": "facebook/sam2.1-hiera-large@665f8e2ad61cf5f53d65644ff27c8ee525124610",
+      "license": "apache-2.0",
+      "files": {
+        "sam2.1_hiera_large.pt": {
+          "size": 898083611,
+          "sha256": "2647878d5dfa5098f2f8649825738a9345572bae2d4350a2468587ece47dd318"
+        },
+        "LICENSE": {
+          "size": 11357,
+          "sha256": "c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4"
+        },
+        "PROVENANCE.txt": {
+          "size": 191,
+          "sha256": "55039cfc30d353c574dd17f2c846537384a8743b0000a42ed98eff36307e37c7"
+        }
+      }
+    },
+    "sam2-small": {
+      "prefix": "sam2-small",
+      "upstream": "facebook/sam2.1-hiera-small@ee5bba1d82bb8749febdf90f45e84b687142ba03",
+      "license": "apache-2.0",
+      "files": {
+        "sam2.1_hiera_small.pt": {
+          "size": 184416285,
+          "sha256": "6d1aa6f30de5c92224f8172114de081d104bbd23dd9dc5c58996f0cad5dc4d38"
+        },
+        "LICENSE": {
+          "size": 11357,
+          "sha256": "c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4"
+        },
+        "PROVENANCE.txt": {
+          "size": 191,
+          "sha256": "5db26cc2ee5597748b419bc1f457fc5f80dbb403354ab617288de3a7995bfd8f"
+        }
+      }
+    },
+    "sam2-tiny": {
+      "prefix": "sam2-tiny",
+      "upstream": "facebook/sam2.1-hiera-tiny@de431c4043854a71d8101e17995dfe596bf101a5",
+      "license": "apache-2.0",
+      "files": {
+        "sam2.1_hiera_tiny.pt": {
+          "size": 156008466,
+          "sha256": "7402e0d864fa82708a20fbd15bc84245c2f26dff0eb43a4b5b93452deb34be69"
+        },
+        "LICENSE": {
+          "size": 11357,
+          "sha256": "c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4"
+        },
+        "PROVENANCE.txt": {
+          "size": 190,
+          "sha256": "f11cc4c485393c9fb1073ca11f9f1494fce9122095f6e17a242685657f147f3e"
+        }
+      }
+    },
+    "tesseract-eng": {
+      "prefix": "tesseract-eng",
+      "upstream": "https://cdn.jsdelivr.net/npm/@tesseract.js-data/eng/4.0.0_best_int/eng.traineddata.gz",
+      "license": "apache-2.0",
+      "files": {
+        "eng.traineddata.gz": {
+          "size": 2952873,
+          "sha256": "45b4cb346724ac1774f1c36f42f182b887bcdb28ebe63e6fff90ac41f3fcff91"
+        },
+        "LICENSE": {
+          "size": 11357,
+          "sha256": "c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4"
+        },
+        "PROVENANCE.txt": {
+          "size": 124,
+          "sha256": "dcfa4baae68e25fd7108136eefd43ac3f7228fe3356f527bc9e67d84ad278a01"
+        }
+      }
+    },
+    "unlimited-ocr": {
+      "prefix": "unlimited-ocr",
+      "upstream": "baidu/Unlimited-OCR@8249d20b2822a7d8a4c22c8d8c9069b5ac36bda3",
+      "license": "mit",
+      "files": {
+        "LICENSE": {
+          "size": 1061,
+          "sha256": "d985048c6d69429d685fdbe7557340aa0897c0fd8dc038299b148b9c75dc3383"
+        },
+        "README.md": {
+          "size": 11108,
+          "sha256": "4f573db255db5262bfe2ea92e459bfc11635a456f210e3a96289bb9605077a55"
+        },
+        "chat_template.jinja": {
+          "size": 529,
+          "sha256": "690dcec206a691b4876d70db7092b6b0ae14cb8b51396e4c89a25b777f5a6830"
+        },
+        "config.json": {
+          "size": 2881,
+          "sha256": "55e6c5179eea054a3fca99e3e41424649d22f574fba4c5f0737ef9c9095526e4"
+        },
+        "configuration_deepseek_v2.py": {
+          "size": 10720,
+          "sha256": "b8470dd616ba8745fce6e27b093aef73a098863cc891b2477dcf9326a36000f7"
+        },
+        "conversation.py": {
+          "size": 9253,
+          "sha256": "ec7b6ce89bcda643de1f43269ffa66a7b2e65dc3ed30e427958f776546b4ba03"
+        },
+        "deepencoder.py": {
+          "size": 38008,
+          "sha256": "0ae2fb6d1e5ae8cf100fc32f854830acd08c821a0a1f23a94a76588c222ddcf2"
+        },
+        "generation_config.json": {
+          "size": 202,
+          "sha256": "dc97b67afaad12b72320c4230f292be8cb5a73fcb4caa9072074fe5a341fb003"
+        },
+        "model-00001-of-000001.safetensors": {
+          "size": 6672547120,
+          "sha256": "2bc48a7a110061ea58fff65d3169367eebe3aee371ca6968dc2219c1b2855fc6"
+        },
+        "model.safetensors.index.json": {
+          "size": 257611,
+          "sha256": "354be1f2dcfb72ebb385e25465522ce5413a77c36f3b35fec088a3162a11af99"
+        },
+        "modeling_deepseekv2.py": {
+          "size": 90162,
+          "sha256": "74e36e6bd0ba7bc565ef76464a99baa8e6bccb710ae9c1007b54ac30b855fa4c"
+        },
+        "modeling_unlimitedocr.py": {
+          "size": 53411,
+          "sha256": "3894696d68197ad00e89f0963880645dfce49b1933a9f27e78fbc05983547a78"
+        },
+        "preprocessor_config.json": {
+          "size": 457,
+          "sha256": "62cef9e2db4ae494ff2ecccaedfeef884d8ba880c25d17f14e81f3d1c3c899cb"
+        },
+        "processor_config.json": {
+          "size": 452,
+          "sha256": "2c82d81788aac5146db01d842138e8f5921daa907d2fd61c3c6c757d41dfb1d7"
+        },
+        "special_tokens_map.json": {
+          "size": 801,
+          "sha256": "ab4bd57ce17d62e39e0a39e739de1e407484f090f0b2c7e391312bca7a5b061a"
+        },
+        "tokenizer.json": {
+          "size": 9979544,
+          "sha256": "a02f8fd5228c90256bb4f6554c34a579d48f909e5beb232dc4afad870b55a8b4"
+        },
+        "tokenizer_config.json": {
+          "size": 165938,
+          "sha256": "a0cbe8464049da1f891b7a12676de06af4cb54c130995d42f71adc1c30c6e9f3"
+        },
+        "PROVENANCE.txt": {
+          "size": 176,
+          "sha256": "6a4046757ec2c1b5978855e536406929e3f6a91113c4ed3c4550883368e3ba1d"
+        }
+      }
+    },
+    "whisper-large-v3": {
+      "prefix": "whisper-large-v3",
+      "upstream": "Systran/faster-whisper-large-v3@edaa852ec7e145841d8ffdb056a99866b5f0a478",
+      "license": "mit",
+      "files": {
+        "README.md": {
+          "size": 2052,
+          "sha256": "39e96252229f5a3d0141dc81afb65a36fd205461ac21e5b70f2cd1248ef0082c"
+        },
+        "config.json": {
+          "size": 2394,
+          "sha256": "a9306624f5ec14270a014b647e5c316b6e03a662c369758d1b90697a7b0655b9"
+        },
+        "model.bin": {
+          "size": 3087284237,
+          "sha256": "69f74147e3334731bc3a76048724833325d2ec74642fb52620eda87352e3d4f1"
+        },
+        "preprocessor_config.json": {
+          "size": 340,
+          "sha256": "7ccc62c6f2765af1f3b46c00c9b5894426835a05021c8b9c01eecb6dfb542711"
+        },
+        "tokenizer.json": {
+          "size": 2480617,
+          "sha256": "6d8cbd7cd0d8d5815e478dac67b85a26bbe77c1f5e0c6d76d1ce2abc0e5f21ca"
+        },
+        "vocabulary.json": {
+          "size": 1068114,
+          "sha256": "c69260f2ab26d659b7c398f9a2b2b48ed0df16c3b47d7326782fd9cba71690c1"
+        },
+        "LICENSE": {
+          "size": 1063,
+          "sha256": "b5d65a59060e68c4ff940e1eddfa6f94b2d68fdf58ed7f4dd57721c997e35e9d"
+        },
+        "PROVENANCE.txt": {
+          "size": 188,
+          "sha256": "c917b3927e21c81a42d4c26d6cefed78a115e05fe272e30b2cd7b605b92a179c"
+        }
+      }
+    },
+    "whisper-small": {
+      "prefix": "whisper-small",
+      "upstream": "Systran/faster-whisper-small@536b0662742c02347bc0e980a01041f333bce120",
+      "license": "mit",
+      "files": {
+        "README.md": {
+          "size": 1998,
+          "sha256": "329373481008c7c38654aff8ecdcf0163c211557cc7ba8e2ef6f2f84b4f75ec8"
+        },
+        "config.json": {
+          "size": 2370,
+          "sha256": "b55496ac7940a7ae47d2c01eab40edfd8701feec1229d9cce3b40014383fb828"
+        },
+        "model.bin": {
+          "size": 483546902,
+          "sha256": "3e305921506d8872816023e4c273e75d2419fb89b24da97b4fe7bce14170d671"
+        },
+        "tokenizer.json": {
+          "size": 2203239,
+          "sha256": "fb7b63191e9bb045082c79fd742a3106a12c99513ab30df4a0d47fa6cb6fd0ab"
+        },
+        "vocabulary.txt": {
+          "size": 459861,
+          "sha256": "34ce3fe1c5041027b3f8d42912270993f986dbc4bb34cf27f951e34a1e453913"
+        },
+        "LICENSE": {
+          "size": 1063,
+          "sha256": "b5d65a59060e68c4ff940e1eddfa6f94b2d68fdf58ed7f4dd57721c997e35e9d"
+        },
+        "PROVENANCE.txt": {
+          "size": 185,
+          "sha256": "3a8afb611fb355e25ca6ddd9a7c9c9bf5526d74f880a6cdde62c1f647edc83a4"
+        }
+      }
+    }
+  }
+}

README.md

@@ -1,47 +1,50 @@
----
-license: other
-license_name: sam
-license_link: LICENSE
-base_model: facebook/sam3
-library_name: transformers
-pipeline_tag: mask-generation
-tags:
-  - sam3
-  - snaptop
-  - onnx
----
-
-# SnapTop model distribution
-
-This is a dedicated model repository for SnapTop. It is separate from the
-application source repository and other repositories owned by this account.
-
-SnapTop downloads these files for local inference. Screenshots, text prompts
-and videos stay on the user's computer. Public downloads require no Hugging
-Face account or access token.
-
-## Contents and provenance
-
-- `model.safetensors` and the processor/tokenizer files: unchanged SAM3
-  Transformers weights from [facebook/sam3](https://huggingface.co/facebook/sam3),
-  revision `3c879f39826c281e95690f02c7821c4de09afae7`.
-- `sam3_vit_h.zip`: unchanged ONNX export from
-  [vietanhdev/segment-anything-3-onnx-models](https://huggingface.co/vietanhdev/segment-anything-3-onnx-models),
-  used by SnapTop's built-in image segmentation engine.
-- `LICENSE`: the complete Meta SAM License accompanying the SAM3 materials.
-- `SHA256SUMS`: checksums of the distribution files.
-
-The weights were developed by Meta, not SnapTop. No fine-tuning or model
-conversion was performed by this repository's maintainer. Redistribution
-does not change the original SAM License or its use restrictions. ONNX
-packaging and conversion attribution belongs to the upstream converter.
-
-## Downloads
-
-SnapTop release builds pin a repository revision and verify downloaded model
-files. Each backend downloads its own format on first use and reuses the local
-cache. Installing the app alone does not download both multi-gigabyte formats.
-
-The Python engine uses the Transformers model; the built-in engine uses the
-ONNX archive. Both formats are hosted here so customers do not need to request
-access to the upstream gated repository.
+---
+license: other
+license_name: model-specific
+license_link: https://huggingface.co/hacksider/snaptop-models/blob/main/LICENSES.md
+tags:
+  - snaptop
+  - sam3
+  - onnx
+  - speech
+  - ocr
+---
+
+# SnapTop model catalog
+
+The dedicated public model repository for SnapTop, separate from application source and all other account repositories.
+All SnapTop model downloads originate here. Files are public; no account, login or access token is needed.
+Models run locally: screenshots, recordings, prompts and voice samples remain on the user's computer.
+
+SnapTop downloads models on demand using the immutable commit and SHA-256 checksums recorded in the application's `sam/model-source.json`.
+`MODEL_CATALOG.json` records file sizes, checksums and exact upstream revisions. `SHA256SUMS` covers all distributed model payloads.
+Each model folder includes its original license and provenance. The licenses are model-specific; this repository does not relicense model weights.
+
+## SAM3
+
+The root `model.safetensors` and processor/tokenizer files are unchanged Meta SAM3 weights from
+[facebook/sam3@3c879f39826c281e95690f02c7821c4de09afae7](https://huggingface.co/facebook/sam3/tree/3c879f39826c281e95690f02c7821c4de09afae7).
+The root `sam3_vit_h.zip` is the unchanged ONNX conversion from
+[vietanhdev/segment-anything-3-onnx-models](https://huggingface.co/vietanhdev/segment-anything-3-onnx-models).
+Both retain the [Meta SAM License](LICENSE). The root Transformers files remain a loadable SAM3 snapshot.
+
+## Additional models
+
+| Model | License | Original source |
+| --- | --- | --- |
+| [chatterbox-turbo](chatterbox-turbo) | [mit](chatterbox-turbo/LICENSE) | [ResembleAI/chatterbox-turbo](https://huggingface.co/ResembleAI/chatterbox-turbo/tree/749d1c1a46eb10492095d68fbcf55691ccf137cd) |
+| [madlad-translation](madlad-translation) | [apache-2.0](madlad-translation/LICENSE) | [google/madlad400-3b-mt](https://huggingface.co/google/madlad400-3b-mt/tree/fa184c675da0b5c9e1c8694fccd4e12e2d422094) |
+| [qwen-aligner](qwen-aligner) | [apache-2.0](qwen-aligner/LICENSE) | [Qwen/Qwen3-ForcedAligner-0.6B](https://huggingface.co/Qwen/Qwen3-ForcedAligner-0.6B/tree/c7cbfc2048c462b0d63a45797104fc9db3ad62b7) |
+| [qwen-asr](qwen-asr) | [apache-2.0](qwen-asr/LICENSE) | [Qwen/Qwen3-ASR-1.7B](https://huggingface.co/Qwen/Qwen3-ASR-1.7B/tree/7278e1e70fe206f11671096ffdd38061171dd6e5) |
+| [rnnoise](rnnoise) | [bsd-3-clause](rnnoise/LICENSE) | [https://raw.githubusercontent.com/richardpl/arnndn-models/0fda24c46d78f0207d820bb970fbe85c1971b39c/std.rnnn](https://raw.githubusercontent.com/richardpl/arnndn-models/0fda24c46d78f0207d820bb970fbe85c1971b39c/std.rnnn) |
+| [sam2-base-plus](sam2-base-plus) | [apache-2.0](sam2-base-plus/LICENSE) | [facebook/sam2.1-hiera-base-plus](https://huggingface.co/facebook/sam2.1-hiera-base-plus/tree/b7320756a13354e7530a63935656d35b2f91a290) |
+| [sam2-large](sam2-large) | [apache-2.0](sam2-large/LICENSE) | [facebook/sam2.1-hiera-large](https://huggingface.co/facebook/sam2.1-hiera-large/tree/665f8e2ad61cf5f53d65644ff27c8ee525124610) |
+| [sam2-small](sam2-small) | [apache-2.0](sam2-small/LICENSE) | [facebook/sam2.1-hiera-small](https://huggingface.co/facebook/sam2.1-hiera-small/tree/ee5bba1d82bb8749febdf90f45e84b687142ba03) |
+| [sam2-tiny](sam2-tiny) | [apache-2.0](sam2-tiny/LICENSE) | [facebook/sam2.1-hiera-tiny](https://huggingface.co/facebook/sam2.1-hiera-tiny/tree/de431c4043854a71d8101e17995dfe596bf101a5) |
+| [tesseract-eng](tesseract-eng) | [apache-2.0](tesseract-eng/LICENSE) | [https://cdn.jsdelivr.net/npm/](https://huggingface.co/https://cdn.jsdelivr.net/npm//tree/tesseract.js-data/eng/4.0.0_best_int/eng.traineddata.gz) |
+| [unlimited-ocr](unlimited-ocr) | [mit](unlimited-ocr/LICENSE) | [baidu/Unlimited-OCR](https://huggingface.co/baidu/Unlimited-OCR/tree/8249d20b2822a7d8a4c22c8d8c9069b5ac36bda3) |
+| [whisper-large-v3](whisper-large-v3) | [mit](whisper-large-v3/LICENSE) | [Systran/faster-whisper-large-v3](https://huggingface.co/Systran/faster-whisper-large-v3/tree/edaa852ec7e145841d8ffdb056a99866b5f0a478) |
+| [whisper-small](whisper-small) | [mit](whisper-small/LICENSE) | [Systran/faster-whisper-small](https://huggingface.co/Systran/faster-whisper-small/tree/536b0662742c02347bc0e980a01041f333bce120) |
+
+MADLAD is used only to generate application translations during development. Qwen ASR/alignment and Unlimited OCR are optional engines.
+Software dependencies and runtimes are installed from their official package sources; they are not model payloads in this repository.

SHA256SUMS

@@ -1,10 +1,107 @@
 bec48f70bd37bf8280a9d1ebf01642d26086f6122ba735baa08fe03c5a6e7448  LICENSE
+4248e910a928849fe5815a0f9236e17fa07768d95b9193212752c464b93d6caa  chatterbox-turbo/LICENSE
+633e03dbd0393ca8fb3ee2eb9a1c5ec6708398380d6e713b7e1a64e71b34e327  chatterbox-turbo/PROVENANCE.txt
+d007d8c2ed2093530d95eee9d623f8977c321f6f361c19dafa07c000375808d7  chatterbox-turbo/README.md
+72e4ab6acb0d9309ac3df4b526ae5fd80a2da5bc5ab7bb02d85096a374f69193  chatterbox-turbo/added_tokens.json
+b1852099306fd6a7814eb9d0bd10186caba7249596cc23868f78a0eefbfa5033  chatterbox-turbo/conds.pt
+1ce1664773c50f3e0cc8842619a93edc4624525b728b188a9e0be33b7726adc5  chatterbox-turbo/merges.txt
+d65cb687a2ed581ee6cc297e919ffefa63386944f42364ae13b78a594945514f  chatterbox-turbo/s3gen_meanflow.safetensors
+92ba8063bf40aa163eadebbfe0de07c2aebe44cf0d4a9e8726580b0781fd2640  chatterbox-turbo/special_tokens_map.json
+fcf1f8c1d651bb7e3acd69ee5be269b4ac10c02980b7708213d598bc9f7cdf87  chatterbox-turbo/t3_turbo_v1.safetensors
+bca16a2ac1ddbd78b8d6228f0031884cc74b6ea54b967d6f6d2ebae9ccde23e6  chatterbox-turbo/tokenizer_config.json
+f0921cab452fa278bc25cd23ffd59d36f816d7dc5181dd1bef9751a7fb61f63c  chatterbox-turbo/ve.safetensors
+f6bd25a65e4e63ca31360e9fb11c7e4f9a391a78385d640acd814092dd6eee4f  chatterbox-turbo/vocab.json
 4616385e4b21f2e5e22c875b65679185cbccfa95de42542b9166f7dc3d57160f  config.json
+c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4  madlad-translation/LICENSE
+9cfafdae9362a042d581da47691054ea8f5eb8aa628d59ab2866e975c3366f44  madlad-translation/PROVENANCE.txt
+d914176fd50bd7f565700006a31aa97b79d3ad17cee20c8e5ff2061d5cb74817  madlad-translation/added_tokens.json
+cad399cab799b99409a6ec2d90d72552257c2bb752861261d2016691e0643e7c  madlad-translation/config.json
+0849b38987568ccfe4ebefc22bbda1cec4bee01c345e93bfab207d4692b0a1d5  madlad-translation/generation_config.json
+66ff5f8fcaf92291da486fdfbd4d5233cec90e1359348a56e3172c978b3a76d4  madlad-translation/model.safetensors
+7f79f1d5063d56c4b980eec0692f3c7429bdef335071d34e566bd00fd4b5e3e0  madlad-translation/special_tokens_map.json
+ef11ac9a22c7503492f56d48dce53be20e339b63605983e9f27d2cd0e0f3922c  madlad-translation/spiece.model
+a2799ccc696b752ba00c34f58726bfe253a04921ceb6cfc620400f560474790b  madlad-translation/tokenizer.json
+641fc660745306dfb935f666a68f8bc10a44c39241cfb357be518fda8c09662d  madlad-translation/tokenizer_config.json
 9fd691f7c8039210e0fced15865466c65820d09b63988b0174bfe25de299051a  merges.txt
 6d06f0a5f84e435071fe6603e61d0b4cc7b40e0d39d487cfd4d67d8cc11cc14a  model.safetensors
 6420cf2671fa9309ea95bc0144a8b9861666d1c5f43c8db09e410dacda974fce  processor_config.json
+c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4  qwen-aligner/LICENSE
+e536da59d661aeee5e2e1f63825af871f0fb510c2272f213597d5d1945b2c154  qwen-aligner/PROVENANCE.txt
+5058416891bc47a2051557765997e8c42f8eb78a0e33c3e775bd17d4b0ba4d50  qwen-aligner/README.md
+75a8cfca24f00de72d796fbfed6858fc9614ef3dabd8696684cc3bc03a9c58ff  qwen-aligner/chat_template.json
+d616c65d46c4b90bdc651b0a0963ea932732241140f337f9bb6b0335a9c8ef09  qwen-aligner/config.json
+948d089b23bca1d214e768d59c4438365665f52ec6d33678f4062206b3fbbb8c  qwen-aligner/generation_config.json
+8831e4f1a044471340f7c0a83d7bd71306a5b867e95fd870f74d0c5308a904d5  qwen-aligner/merges.txt
+47831d0e82f96b20e9034dba01a075ee06436654719f6a68289e49f1b65ce0e7  qwen-aligner/model.safetensors
+45e120a4eda2c20c5d7f2ea9354e63536bf35e27aa573fb7cdf78017b378770d  qwen-aligner/preprocessor_config.json
+3ab80063f8511deb9566e6ad438d17b7a6277fcffd52d92854112f19d36bd81c  qwen-aligner/tokenizer_config.json
+ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910  qwen-aligner/vocab.json
+c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4  qwen-asr/LICENSE
+cfcbdf6e015bf22a5580bbed4e314d8053ad13b880ec1c620700c6aa5f18804d  qwen-asr/PROVENANCE.txt
+5058416891bc47a2051557765997e8c42f8eb78a0e33c3e775bd17d4b0ba4d50  qwen-asr/README.md
+75a8cfca24f00de72d796fbfed6858fc9614ef3dabd8696684cc3bc03a9c58ff  qwen-asr/chat_template.json
+2e74a751548b8ad7d7526d29365ad8144c345d8b412b1152d25dc6698452712f  qwen-asr/config.json
+1da527824d81e07118facff437e03f2e24a23311e3bdeb2368973fe77e5f275c  qwen-asr/generation_config.json
+8831e4f1a044471340f7c0a83d7bd71306a5b867e95fd870f74d0c5308a904d5  qwen-asr/merges.txt
+a4cd1f1a04d90b757dc7f7dd26254e69a013b19e80efe590a83c6a3bde8608d6  qwen-asr/model-00001-of-00002.safetensors
+6e0b9d9e09e2e0238e7ef3cc8a484ab387e91b90f1900bedf88bc92d7929ccfc  qwen-asr/model-00002-of-00002.safetensors
+f994739fe38e5210b9e3e8ce6c6307315e2ceac3cb630e7b7414d69dce520f60  qwen-asr/model.safetensors.index.json
+45e120a4eda2c20c5d7f2ea9354e63536bf35e27aa573fb7cdf78017b378770d  qwen-asr/preprocessor_config.json
+4942d005604266809309cabc9f4e9cb89ce855d59b14681fdc0e1cc62ea26c4c  qwen-asr/tokenizer_config.json
+ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910  qwen-asr/vocab.json
+e2f59ff41d9d03adc3dcf3deff170f8c8cf4a6eb4a9b174762a7656d23200ffa  rnnoise/LICENSE
+81933a7bcf5c44b242332e3adafa6140d735c4309e959bb7ea1038156a3ffb6c  rnnoise/PROVENANCE.txt
+6b8943dc4a9b6b24425873992a44f29c0577503276456af46a8854774faeb294  rnnoise/std.rnnn
+c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4  sam2-base-plus/LICENSE
+effcb3a166b276d98dc828e6aede623731ad214ec78fb78dffe8f46ab4b4c979  sam2-base-plus/PROVENANCE.txt
+a2345aede8715ab1d5d31b4a509fb160c5a4af1970f199d9054ccfb746c004c5  sam2-base-plus/sam2.1_hiera_base_plus.pt
+c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4  sam2-large/LICENSE
+55039cfc30d353c574dd17f2c846537384a8743b0000a42ed98eff36307e37c7  sam2-large/PROVENANCE.txt
+2647878d5dfa5098f2f8649825738a9345572bae2d4350a2468587ece47dd318  sam2-large/sam2.1_hiera_large.pt
+c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4  sam2-small/LICENSE
+5db26cc2ee5597748b419bc1f457fc5f80dbb403354ab617288de3a7995bfd8f  sam2-small/PROVENANCE.txt
+6d1aa6f30de5c92224f8172114de081d104bbd23dd9dc5c58996f0cad5dc4d38  sam2-small/sam2.1_hiera_small.pt
+c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4  sam2-tiny/LICENSE
+f11cc4c485393c9fb1073ca11f9f1494fce9122095f6e17a242685657f147f3e  sam2-tiny/PROVENANCE.txt
+7402e0d864fa82708a20fbd15bc84245c2f26dff0eb43a4b5b93452deb34be69  sam2-tiny/sam2.1_hiera_tiny.pt
+1617436f9db7eb9dd69029c2c465d54b316c36c21143f0fde0d1be513f15b7d5  sam3_vit_h.zip
 2cdb3b8331a60c92fc1e55a13e9fd61fd2293c5a51275fdcccd62b780052530e  special_tokens_map.json
+c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4  tesseract-eng/LICENSE
+dcfa4baae68e25fd7108136eefd43ac3f7228fe3356f527bc9e67d84ad278a01  tesseract-eng/PROVENANCE.txt
+45b4cb346724ac1774f1c36f42f182b887bcdb28ebe63e6fff90ac41f3fcff91  tesseract-eng/eng.traineddata.gz
 6d9109cc838977f3ca94a379eec36aecc7c807e1785cd729660ca2fc0171fb35  tokenizer.json
 39670ad98457fe8f14ca59f6bb74591e9fc850974380a63993e5b8ffc865baa2  tokenizer_config.json
+d985048c6d69429d685fdbe7557340aa0897c0fd8dc038299b148b9c75dc3383  unlimited-ocr/LICENSE
+6a4046757ec2c1b5978855e536406929e3f6a91113c4ed3c4550883368e3ba1d  unlimited-ocr/PROVENANCE.txt
+4f573db255db5262bfe2ea92e459bfc11635a456f210e3a96289bb9605077a55  unlimited-ocr/README.md
+690dcec206a691b4876d70db7092b6b0ae14cb8b51396e4c89a25b777f5a6830  unlimited-ocr/chat_template.jinja
+55e6c5179eea054a3fca99e3e41424649d22f574fba4c5f0737ef9c9095526e4  unlimited-ocr/config.json
+b8470dd616ba8745fce6e27b093aef73a098863cc891b2477dcf9326a36000f7  unlimited-ocr/configuration_deepseek_v2.py
+ec7b6ce89bcda643de1f43269ffa66a7b2e65dc3ed30e427958f776546b4ba03  unlimited-ocr/conversation.py
+0ae2fb6d1e5ae8cf100fc32f854830acd08c821a0a1f23a94a76588c222ddcf2  unlimited-ocr/deepencoder.py
+dc97b67afaad12b72320c4230f292be8cb5a73fcb4caa9072074fe5a341fb003  unlimited-ocr/generation_config.json
+2bc48a7a110061ea58fff65d3169367eebe3aee371ca6968dc2219c1b2855fc6  unlimited-ocr/model-00001-of-000001.safetensors
+354be1f2dcfb72ebb385e25465522ce5413a77c36f3b35fec088a3162a11af99  unlimited-ocr/model.safetensors.index.json
+74e36e6bd0ba7bc565ef76464a99baa8e6bccb710ae9c1007b54ac30b855fa4c  unlimited-ocr/modeling_deepseekv2.py
+3894696d68197ad00e89f0963880645dfce49b1933a9f27e78fbc05983547a78  unlimited-ocr/modeling_unlimitedocr.py
+62cef9e2db4ae494ff2ecccaedfeef884d8ba880c25d17f14e81f3d1c3c899cb  unlimited-ocr/preprocessor_config.json
+2c82d81788aac5146db01d842138e8f5921daa907d2fd61c3c6c757d41dfb1d7  unlimited-ocr/processor_config.json
+ab4bd57ce17d62e39e0a39e739de1e407484f090f0b2c7e391312bca7a5b061a  unlimited-ocr/special_tokens_map.json
+a02f8fd5228c90256bb4f6554c34a579d48f909e5beb232dc4afad870b55a8b4  unlimited-ocr/tokenizer.json
+a0cbe8464049da1f891b7a12676de06af4cb54c130995d42f71adc1c30c6e9f3  unlimited-ocr/tokenizer_config.json
 5047b556ce86ccaf6aa22b3ffccfc52d391ea4accdab9c2f2407da5b742d4363  vocab.json
-1617436f9db7eb9dd69029c2c465d54b316c36c21143f0fde0d1be513f15b7d5  sam3_vit_h.zip
+b5d65a59060e68c4ff940e1eddfa6f94b2d68fdf58ed7f4dd57721c997e35e9d  whisper-large-v3/LICENSE
+c917b3927e21c81a42d4c26d6cefed78a115e05fe272e30b2cd7b605b92a179c  whisper-large-v3/PROVENANCE.txt
+39e96252229f5a3d0141dc81afb65a36fd205461ac21e5b70f2cd1248ef0082c  whisper-large-v3/README.md
+a9306624f5ec14270a014b647e5c316b6e03a662c369758d1b90697a7b0655b9  whisper-large-v3/config.json
+69f74147e3334731bc3a76048724833325d2ec74642fb52620eda87352e3d4f1  whisper-large-v3/model.bin
+7ccc62c6f2765af1f3b46c00c9b5894426835a05021c8b9c01eecb6dfb542711  whisper-large-v3/preprocessor_config.json
+6d8cbd7cd0d8d5815e478dac67b85a26bbe77c1f5e0c6d76d1ce2abc0e5f21ca  whisper-large-v3/tokenizer.json
+c69260f2ab26d659b7c398f9a2b2b48ed0df16c3b47d7326782fd9cba71690c1  whisper-large-v3/vocabulary.json
+b5d65a59060e68c4ff940e1eddfa6f94b2d68fdf58ed7f4dd57721c997e35e9d  whisper-small/LICENSE
+3a8afb611fb355e25ca6ddd9a7c9c9bf5526d74f880a6cdde62c1f647edc83a4  whisper-small/PROVENANCE.txt
+329373481008c7c38654aff8ecdcf0163c211557cc7ba8e2ef6f2f84b4f75ec8  whisper-small/README.md
+b55496ac7940a7ae47d2c01eab40edfd8701feec1229d9cce3b40014383fb828  whisper-small/config.json
+3e305921506d8872816023e4c273e75d2419fb89b24da97b4fe7bce14170d671  whisper-small/model.bin
+fb7b63191e9bb045082c79fd742a3106a12c99513ab30df4a0d47fa6cb6fd0ab  whisper-small/tokenizer.json
+34ce3fe1c5041027b3f8d42912270993f986dbc4bb34cf27f951e34a1e453913  whisper-small/vocabulary.txt

madlad-translation/LICENSE

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

madlad-translation/PROVENANCE.txt

@@ -0,0 +1,4 @@
+Original model: google/madlad400-3b-mt
+Upstream revision: fa184c675da0b5c9e1c8694fccd4e12e2d422094
+License: apache-2.0
+Distributed for SnapTop without changing the upstream license.

madlad-translation/added_tokens.json

@@ -0,0 +1,2 @@
+{
+}

madlad-translation/config.json

@@ -0,0 +1,29 @@
+{
+    "architectures": [
+        "T5ForConditionalGeneration"
+    ],
+    "d_ff": 8192,
+    "d_kv": 128,
+    "d_model": 1024,
+    "dropout_rate": 0.1,
+    "decoder_start_token_id": 0,
+    "pad_token_id": 1,
+    "eos_token_id": 2,
+    "feed_forward_proj": "gated-gelu",
+    "initializer_factor": 1.0,
+    "is_encoder_decoder": true,
+    "layer_norm_epsilon": 1e-06,
+    "model_type": "t5",
+    "n_positions": 512,
+    "num_decoder_layers": 32,
+    "num_heads": 16,
+    "num_layers": 32,
+    "output_past": true,
+    "relative_attention_max_distance": 128,
+    "relative_attention_num_buckets": 32,
+    "task_specific_params": {},
+    "tie_word_embeddings": false,
+    "transformers_version": "4.23.1",
+    "use_cache": true,
+    "vocab_size": 256000
+}

madlad-translation/generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "decoder_start_token_id": 0,
+  "eos_token_id": 2,
+  "pad_token_id": 1,
+  "transformers_version": "4.35.0"
+}

madlad-translation/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:66ff5f8fcaf92291da486fdfbd4d5233cec90e1359348a56e3172c978b3a76d4
+size 11761587872

madlad-translation/special_tokens_map.json

@@ -0,0 +1,23 @@
+{
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

madlad-translation/spiece.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef11ac9a22c7503492f56d48dce53be20e339b63605983e9f27d2cd0e0f3922c
+size 4427844

madlad-translation/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2799ccc696b752ba00c34f58726bfe253a04921ceb6cfc620400f560474790b
+size 16629031

madlad-translation/tokenizer_config.json

@@ -0,0 +1,38 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [],
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "</s>",
+  "extra_ids": 0,
+  "legacy": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<s>",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "T5Tokenizer",
+  "unk_token": "<unk>"
+}

qwen-aligner/LICENSE

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

qwen-aligner/PROVENANCE.txt

@@ -0,0 +1,4 @@
+Original model: Qwen/Qwen3-ForcedAligner-0.6B
+Upstream revision: c7cbfc2048c462b0d63a45797104fc9db3ad62b7
+License: apache-2.0
+Distributed for SnapTop without changing the upstream license.

qwen-aligner/README.md

@@ -0,0 +1,1393 @@
+---
+license: apache-2.0
+pipeline_tag: automatic-speech-recognition
+---
+
+# Qwen3-ASR
+
+## Overview
+
+### Introduction
+
+<p align="center">
+    <img src="https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-ASR-Repo/qwen3_asr_introduction.png" width="90%"/>
+<p>
+
+The Qwen3-ASR family includes Qwen3-ASR-1.7B and Qwen3-ASR-0.6B, which support language identification and ASR for 52 languages and dialects. Both leverage large-scale speech training data and the strong audio understanding capability of their foundation model, Qwen3-Omni. Experiments show that the 1.7B version achieves state-of-the-art performance among open-source ASR models and is competitive with the strongest proprietary commercial APIs. Here are the main features:
+
+* **All-in-one**: Qwen3-ASR-1.7B and Qwen3-ASR-0.6B support language identification and speech recognition for 30 languages and 22 Chinese dialects, so as to English accents from multiple countries and regions.
+
+* **Excellent and Fast**: The Qwen3-ASR family ASR models maintains high-quality and robust recognition under complex acoustic environments and challenging text patterns. Qwen3-ASR-1.7B achieves strong performance on both open-sourced and internal benchmarks. While the 0.6B version achieves accuracy-efficient trade-off, it reaches 2000 times throughput at a concurrency of 128. They both achieve streaming / offline unified inference with single model and support transcribe long audio.
+
+* **Novel and strong forced alignment Solution**: We introduce Qwen3-ForcedAligner-0.6B, which supports timestamp prediction for arbitrary units within up to 5 minutes of speech in 11 languages. Evaluations show its timestamp accuracy surpasses E2E based forced-alignment models.
+
+* **Comprehensive inference toolkit**: In addition to open-sourcing the architectures and weights of the Qwen3-ASR series, we also release a powerful, full-featured inference framework that supports vLLM-based batch inference, asynchronous serving, streaming inference, timestamp prediction, and more.
+
+### Model Architecture
+
+<p align="center">
+    <img src="https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-ASR-Repo/overview.jpg" width="100%"/>
+<p>
+
+
+### Released Models Description and Download
+
+Below is an introduction and download information for the Qwen3-ASR models. Please select and download the model that fits your needs.
+
+| Model | Supported Languages | Supported Dialects | Inference Mode | Audio Types |
+|---|---|---|---|---|
+| Qwen3-ASR-1.7B & Qwen3-ASR-0.6B | Chinese (zh), English (en), Cantonese (yue), Arabic (ar), German (de), French (fr), Spanish (es), Portuguese (pt), Indonesian (id), Italian (it), Korean (ko), Russian (ru), Thai (th), Vietnamese (vi), Japanese (ja), Turkish (tr), Hindi (hi), Malay (ms), Dutch (nl), Swedish (sv), Danish (da), Finnish (fi), Polish (pl), Czech (cs), Filipino (fil), Persian (fa), Greek (el), Hungarian (hu), Macedonian (mk), Romanian (ro) | Anhui, Dongbei, Fujian, Gansu, Guizhou, Hebei, Henan, Hubei, Hunan, Jiangxi, Ningxia, Shandong, Shaanxi, Shanxi, Sichuan, Tianjin, Yunnan, Zhejiang, Cantonese (Hong Kong accent), Cantonese (Guangdong accent), Wu language, Minnan language. | Offline / Streaming | Speech, Singing Voice, Songs with BGM |
+| Qwen3-ForcedAligner-0.6B | Chinese, English, Cantonese, French, German, Italian, Japanese, Korean, Portuguese, Russian, Spanish | -- | NAR | Speech |
+
+During model loading in the `qwen-asr` package or vLLM, model weights will be downloaded automatically based on the model name. However, if your runtime environment does not allow downloading weights during execution, you can use the following commands to manually download the model weights to a local directory:
+
+```bash
+# Download through ModelScope (recommended for users in Mainland China)
+pip install -U modelscope
+modelscope download --model Qwen/Qwen3-ASR-1.7B  --local_dir ./Qwen3-ASR-1.7B
+modelscope download --model Qwen/Qwen3-ASR-0.6B --local_dir ./Qwen3-ASR-0.6B
+modelscope download --model Qwen/Qwen3-ForcedAligner-0.6B --local_dir ./Qwen3-ForcedAligner-0.6B
+# Download through Hugging Face
+pip install -U "huggingface_hub[cli]"
+huggingface-cli download Qwen/Qwen3-ASR-1.7B --local-dir ./Qwen3-ASR-1.7B
+huggingface-cli download Qwen/Qwen3-ASR-0.6B --local-dir ./Qwen3-ASR-0.6B
+huggingface-cli download Qwen/Qwen3-ForcedAligner-0.6B --local-dir ./Qwen3-ForcedAligner-0.6B
+```
+
+
+## Quickstart
+
+### Environment Setup
+
+The easiest way to use Qwen3-ASR is to install the `qwen-asr` Python package from PyPI. This will pull in the required runtime dependencies and allow you to load any released Qwen3-ASR model. If you’d like to simplify environment setup further, you can also use our official [Docker image](#docker). The `qwen-asr` package provides two backends: the transformers backend and the vLLM backend. For usage instructions for different backends, please refer to [Python Package Usage](#python-package-usage). We recommend using a **fresh, isolated environment** to avoid dependency conflicts with existing packages. You can create a clean Python 3.12 environment like this:
+
+```bash
+conda create -n qwen3-asr python=3.12 -y
+conda activate qwen3-asr
+```
+
+Run the following command to get the minimal installation with transformers-backend support:
+
+```bash
+pip install -U qwen-asr
+```
+
+To enable the vLLM backend for faster inference and streaming support, run:
+
+```bash
+pip install -U qwen-asr[vllm]
+```
+
+If you want to develop or modify the code locally, install from source in editable mode:
+
+```bash
+git clone https://github.com/QwenLM/Qwen3-ASR.git
+cd Qwen3-ASR
+pip install -e .
+# support vLLM backend
+# pip install -e ".[vllm]"
+```
+
+Additionally, we recommend using FlashAttention 2 to reduce GPU memory usage and accelerate inference speed, especially for long inputs and large batch sizes.
+
+```bash
+pip install -U flash-attn --no-build-isolation
+```
+
+If your machine has less than 96GB of RAM and lots of CPU cores, run:
+
+```bash
+MAX_JOBS=4 pip install -U flash-attn --no-build-isolation
+```
+
+Also, you should have hardware that is compatible with FlashAttention 2. Read more about it in the official documentation of the [FlashAttention repository](https://github.com/Dao-AILab/flash-attention). FlashAttention 2 can only be used when a model is loaded in `torch.float16` or `torch.bfloat16`.
+
+### Python Package Usage
+
+#### Quick Inference
+
+The `qwen-asr` package provides two backends: **transformers backend** and **vLLM backend**. You can pass audio inputs as a local path, a URL, base64 data, or a `(np.ndarray, sr)` tuple, and run batch inference. To quickly try Qwen3-ASR, you can use `Qwen3ASRModel.from_pretrained(...)` for the transformers backend with the following code:
+
+```python
+import torch
+from qwen_asr import Qwen3ASRModel
+
+model = Qwen3ASRModel.from_pretrained(
+    "Qwen/Qwen3-ASR-1.7B",
+    dtype=torch.bfloat16,
+    device_map="cuda:0",
+    # attn_implementation="flash_attention_2",
+    max_inference_batch_size=32, # Batch size limit for inference. -1 means unlimited. Smaller values can help avoid OOM.
+    max_new_tokens=256, # Maximum number of tokens to generate. Set a larger value for long audio input.
+)
+
+results = model.transcribe(
+    audio="https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-ASR-Repo/asr_en.wav",
+    language=None, # set "English" to force the language
+)
+
+print(results[0].language)
+print(results[0].text)
+```
+
+If you want to return timestamps, pass `forced_aligner` and its init kwargs. Here is an example of batch inference with timestamps output:
+
+```python
+import torch
+from qwen_asr import Qwen3ASRModel
+
+model = Qwen3ASRModel.from_pretrained(
+    "Qwen/Qwen3-ASR-1.7B",
+    dtype=torch.bfloat16,
+    device_map="cuda:0",
+    # attn_implementation="flash_attention_2",
+    max_inference_batch_size=32, # Batch size limit for inference. -1 means unlimited. Smaller values can help avoid OOM.
+    max_new_tokens=256, # Maximum number of tokens to generate. Set a larger value for long audio input.
+    forced_aligner="Qwen/Qwen3-ForcedAligner-0.6B",
+    forced_aligner_kwargs=dict(
+        dtype=torch.bfloat16,
+        device_map="cuda:0",
+        # attn_implementation="flash_attention_2",
+    ),
+)
+
+results = model.transcribe(
+    audio=[
+      "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-ASR-Repo/asr_zh.wav",
+      "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-ASR-Repo/asr_en.wav",
+    ],
+    language=["Chinese", "English"], # can also be set to None for automatic language detection
+    return_time_stamps=True,
+)
+
+for r in results:
+    print(r.language, r.text, r.time_stamps[0])
+```
+
+For more detailed usage examples, please refer to the [example code](https://github.com/QwenLM/Qwen3-ASR/blob/main/examples/example_qwen3_asr_transformers.py) for the transformers backend.
+
+#### vLLM Backend
+
+If you want the fastest inference speed with Qwen3-ASR, we strongly recommend using the vLLM backend by initializing the model with `Qwen3ASRModel.LLM(...)`. Example code is provided below. Note that you must install it via `pip install -U qwen-asr[vllm]`. If you want the model to output timestamps, it’s best to install FlashAttention via `pip install -U flash-attn --no-build-isolation` to speed up inference for the forced aligner model. Remember to wrap your code under `if __name__ == '__main__':` to avoid the `spawn` error described in [vLLM Troubleshooting](https://docs.vllm.ai/en/latest/usage/troubleshooting/#python-multiprocessing).
+
+```python
+import torch
+from qwen_asr import Qwen3ASRModel
+
+if __name__ == '__main__':
+    model = Qwen3ASRModel.LLM(
+        model="Qwen/Qwen3-ASR-1.7B",
+        gpu_memory_utilization=0.7,
+        max_inference_batch_size=128, # Batch size limit for inference. -1 means unlimited. Smaller values can help avoid OOM.
+        max_new_tokens=4096, # Maximum number of tokens to generate. Set a larger value for long audio input.
+        forced_aligner="Qwen/Qwen3-ForcedAligner-0.6B",
+        forced_aligner_kwargs=dict(
+            dtype=torch.bfloat16,
+            device_map="cuda:0",
+            # attn_implementation="flash_attention_2",
+        ),
+    )
+
+    results = model.transcribe(
+        audio=[
+        "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-ASR-Repo/asr_zh.wav",
+        "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-ASR-Repo/asr_en.wav",
+        ],
+        language=["Chinese", "English"], # can also be set to None for automatic language detection
+        return_time_stamps=True,
+    )
+
+    for r in results:
+        print(r.language, r.text, r.time_stamps[0])
+```
+
+For more detailed usage examples, please refer to the [example code](https://github.com/QwenLM/Qwen3-ASR/blob/main/examples/example_qwen3_asr_vllm.py) for the vLLM backend. In addition, you can start a vLLM server via the `qwen-asr-serve` command, which is a wrapper around `vllm serve`. You can pass any arguments supported by `vllm serve`, for example:
+
+```bash
+qwen-asr-serve Qwen/Qwen3-ASR-1.7B --gpu-memory-utilization 0.8 --host 0.0.0.0 --port 8000
+```
+
+And send requests to the server via:
+
+```python
+import requests
+
+url = "http://localhost:8000/v1/chat/completions"
+headers = {"Content-Type": "application/json"}
+
+data = {
+    "messages": [
+        {
+            "role": "user",
+            "content": [
+                {
+                    "type": "audio_url",
+                    "audio_url": {
+                        "url": "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-ASR-Repo/asr_en.wav"
+                    },
+                }
+            ],
+        }
+    ]
+}
+
+response = requests.post(url, headers=headers, json=data, timeout=300)
+response.raise_for_status()
+content = response.json()['choices'][0]['message']['content']
+print(content)
+
+# parse ASR output if you want
+from qwen_asr import parse_asr_output
+language, text = parse_asr_output(content)
+print(language)
+print(text)
+```
+
+#### Streaming Inference
+
+Qwen3-ASR fully supports streaming inference. Currently, streaming inference is only available with the vLLM backend. Note that streaming inference does not support batch inference or returning timestamps. Please refer to the [example code](https://github.com/QwenLM/Qwen3-ASR/blob/main/examples/example_qwen3_asr_vllm_streaming.py) for details. You can also launch a streaming web demo through the [guide](#streaming-demo) to experience Qwen3-ASR’s streaming transcription capabilities. 
+
+#### ForcedAligner Usage
+
+`Qwen3-ForcedAligner-0.6B` can align text–speech pairs and return word or character level timestamps. Here is an example of using the forced aligner directly:
+
+```python
+import torch
+from qwen_asr import Qwen3ForcedAligner
+
+model = Qwen3ForcedAligner.from_pretrained(
+    "Qwen/Qwen3-ForcedAligner-0.6B",
+    dtype=torch.bfloat16,
+    device_map="cuda:0",
+    # attn_implementation="flash_attention_2",
+)
+
+results = model.align(
+    audio="https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-ASR-Repo/asr_zh.wav",
+    text="甚至出现交易几乎停滞的情况。",
+    language="Chinese",
+)
+
+print(results[0])
+print(results[0][0].text, results[0][0].start_time, results[0][0].end_time)
+```
+
+In addition, the forced aligner supports local paths / URLs / base64 data / `(np.ndarray, sr)` inputs and batch inference. Please refer to the [example code](https://github.com/QwenLM/Qwen3-ASR/blob/main/examples/example_qwen3_forced_aligner.py) for details.
+
+### DashScope API Usage
+
+To further explore Qwen3-ASR, we encourage you to try our DashScope API for a faster and more efficient experience. For detailed API information and documentation, please refer to the following:
+
+| API Description | API Documentation (Mainland China) | API Documentation (International) |
+|------------------|-----------------------------------|------------------------------------|
+| Real-time API for Qwen3-ASR. | [https://help.aliyun.com/zh/model-studio/qwen-real-time-speech-recognition](https://help.aliyun.com/zh/model-studio/qwen-real-time-speech-recognition) | [https://www.alibabacloud.com/help/en/model-studio/qwen-real-time-speech-recognition](https://www.alibabacloud.com/help/en/model-studio/qwen-real-time-speech-recognition) |
+| FileTrans API for Qwen3-ASR. | [https://help.aliyun.com/zh/model-studio/qwen-speech-recognition](https://help.aliyun.com/zh/model-studio/qwen-speech-recognition) | [https://www.alibabacloud.com/help/en/model-studio/qwen-speech-recognition](https://www.alibabacloud.com/help/en/model-studio/qwen-speech-recognition) |
+
+
+## Launch Local Web UI Demo
+
+### Gradio Demo
+
+To launch the Qwen3-ASR web UI gradio demo, install the `qwen-asr` package and run `qwen-asr-demo`. Use the command below for help:
+
+```bash
+qwen-asr-demo --help
+```
+
+To launch the demo, you can use the following commands:
+
+```bash
+# Transformers backend
+qwen-asr-demo \
+  --asr-checkpoint Qwen/Qwen3-ASR-1.7B \
+  --backend transformers \
+  --cuda-visible-devices 0 \
+  --ip 0.0.0.0 --port 8000
+
+# Transformers backend + Forced Aligner (enable timestamps)
+qwen-asr-demo \
+  --asr-checkpoint Qwen/Qwen3-ASR-1.7B \
+  --aligner-checkpoint Qwen/Qwen3-ForcedAligner-0.6B \
+  --backend transformers \
+  --cuda-visible-devices 0 \
+  --backend-kwargs '{"device_map":"cuda:0","dtype":"bfloat16","max_inference_batch_size":8,"max_new_tokens":256}' \
+  --aligner-kwargs '{"device_map":"cuda:0","dtype":"bfloat16"}' \
+  --ip 0.0.0.0 --port 8000
+
+# vLLM backend + Forced Aligner (enable timestamps)
+qwen-asr-demo \
+  --asr-checkpoint Qwen/Qwen3-ASR-1.7B \
+  --aligner-checkpoint Qwen/Qwen3-ForcedAligner-0.6B \
+  --backend vllm \
+  --cuda-visible-devices 0 \
+  --backend-kwargs '{"gpu_memory_utilization":0.7,"max_inference_batch_size":8,"max_new_tokens":2048}' \
+  --aligner-kwargs '{"device_map":"cuda:0","dtype":"bfloat16"}' \
+  --ip 0.0.0.0 --port 8000
+```
+
+Then open `http://<your-ip>:8000`, or access it via port forwarding in tools like VS Code.
+
+#### Backend Notes
+
+This demo supports two backends: transformers and vLLM. All backend-specific initialization parameters should be passed via `--backend-kwargs` as a JSON dict. If not provided, the demo will use sensible defaults.
+
+```bash
+# Example: override transformers init args without flash attention
+--backend-kwargs '{"device_map":"cuda:0","dtype":"bfloat16"}'
+
+# Example: override vLLM init args with 65% GPU memory
+--backend-kwargs '{"gpu_memory_utilization":0.65}'
+```
+
+#### CUDA Device Notes
+
+Because vLLM does not follow `cuda:0` style device selection, this demo selects GPUs by setting `CUDA_VISIBLE_DEVICES` via `--cuda-visible-devices`.
+
+```bash
+# Use GPU 0
+--cuda-visible-devices 0
+
+# Use GPU 1
+--cuda-visible-devices 1
+```
+
+#### Timestamps Notes
+
+Timestamps are only available when `--aligner-checkpoint` is provided. If you launch the demo without a forced aligner, the timestamps UI will be hidden automatically.
+
+```bash
+# No forced aligner
+qwen-asr-demo --asr-checkpoint Qwen/Qwen3-ASR-1.7B
+
+# With forced aligner
+qwen-asr-demo \
+  --asr-checkpoint Qwen/Qwen3-ASR-1.7B \
+  --aligner-checkpoint Qwen/Qwen3-ForcedAligner-0.6B
+```
+
+#### HTTPS Notes
+
+To avoid browser microphone permission issues after deploying the server, it is recommended/required to run the gradio service over HTTPS (especially when accessed remotely or behind modern browsers/gateways). Use `--ssl-certfile` and `--ssl-keyfile` to enable HTTPS. First, generate a private key and a self-signed certificate (valid for 365 days):
+
+```bash
+openssl req -x509 -newkey rsa:2048 \
+  -keyout key.pem -out cert.pem \
+  -days 365 -nodes \
+  -subj "/CN=localhost"
+```
+
+Then run the demo with HTTPS:
+
+```bash
+qwen-asr-demo \
+  --asr-checkpoint Qwen/Qwen3-ASR-1.7B \
+  --backend transformers \
+  --cuda-visible-devices 0 \
+  --ip 0.0.0.0 --port 8000 \
+  --ssl-certfile cert.pem \
+  --ssl-keyfile key.pem \
+  --no-ssl-verify
+```
+
+Then open `https://<your-ip>:8000` to use it. If your browser shows a warning, that’s expected for self-signed certificates. For production, use a real certificate.
+
+### Streaming Demo
+
+To experience Qwen3-ASR’s streaming transcription capability in a web UI, we provide a minimal Flask-based streaming demo. The demo captures microphone audio in the browser, resamples it to 16,000 Hz, and continuously pushes PCM chunks to the model. Run the demo with the following command:
+
+```bash
+qwen-asr-demo-streaming \
+  --asr-model-path Qwen/Qwen3-ASR-1.7B \
+  --host 0.0.0.0 \
+  --port 8000 \
+  --gpu-memory-utilization 0.9
+```
+
+Then open `http://<your-ip>:8000`, or access it via port forwarding in tools like VS Code.
+
+## Deployment with vLLM
+
+vLLM officially provides day-0 model support for Qwen3-ASR for efficient inference. 
+
+### Installation
+You can run Qwen3-ASR with vLLM nightly wheel or docker image. To install the nightly version of vLLM, we recommend using `uv` as the environment manager
+```bash
+uv venv
+source .venv/bin/activate
+uv pip install -U vllm --pre \
+    --extra-index-url https://wheels.vllm.ai/nightly/cu129 \
+    --extra-index-url https://download.pytorch.org/whl/cu129 \
+    --index-strategy unsafe-best-match
+uv pip install "vllm[audio]" # For additional audio dependencies
+```
+
+### Online Serving
+You can easily deploy Qwen3-ASR with vLLM by running the following command
+```bash
+vllm serve Qwen/Qwen3-ASR-1.7B
+```
+After the model server is successfully deployed, you can interact with it in multiple ways.
+
+#### Using OpenAI SDK
+```python
+import base64
+import httpx
+from openai import OpenAI
+
+# Initialize client
+client = OpenAI(
+    base_url="http://localhost:8000/v1",
+    api_key="EMPTY"
+)
+
+# Create multimodal chat completion request
+response = client.chat.completions.create(
+    model="Qwen/Qwen3-ASR-1.7B",
+    messages=[
+        {
+            "role": "user",
+            "content": [
+                {
+                    "type": "audio_url",
+                    "audio_url": {
+                        {"url": "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-ASR-Repo/asr_en.wav"}
+                    }
+                }
+            ]
+        }
+    ],
+)
+
+print(response.choices[0].message.content)
+```
+This model is also supported on vLLM with OpenAI transcription API.
+```python
+import httpx
+from openai import OpenAI
+
+# Initialize client
+client = OpenAI(
+    base_url="http://localhost:8000/v1",
+    api_key="EMPTY"
+)
+audio_url = "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-ASR-Repo/asr_en.wav"
+audio_file = httpx.get(audio_url).content
+
+transcription = client.audio.transcriptions.create(
+    model="Qwen/Qwen3-ASR-1.7B",
+    file=audio_file,
+)
+
+print(transcription.text)
+```
+
+#### Using cURL
+```bash
+curl http://localhost:8000/v1/chat/completions \
+    -H "Content-Type: application/json" \
+    -d '{
+    "messages": [
+    {"role": "user", "content": [
+        {"type": "audio_url", "audio_url": {"url": "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-ASR-Repo/asr_en.wav"}}
+    ]}
+    ]
+    }'
+```
+
+### Offline Inference
+See the following example on using vLLM to run offline inference with Qwen3-ASR
+```python
+from vllm import LLM, SamplingParams
+from vllm.assets.audio import AudioAsset
+import base64
+import requests
+
+# Initialize the LLM
+llm = LLM(
+    model="Qwen/Qwen3-ASR-1.7B"
+)
+
+# Load audio
+audio_asset = AudioAsset("winning_call")
+
+# Create conversation with audio content
+conversation = [
+    {
+        "role": "user",
+        "content": [
+            {
+                "type": "audio_url",
+                "audio_url": {"url": audio_asset.url}
+            }
+        ]
+    }
+]
+
+sampling_params = SamplingParams(temperature=0.01, max_tokens=256)
+
+# Run inference using .chat()
+outputs = llm.chat(conversation, sampling_params=sampling_params)
+print(outputs[0].outputs[0].text)
+```
+
+
+## Docker
+
+To make it easier to use our `qwen-asr` Python package, we provide a pre-built Docker image: [qwenllm/qwen3-asr](https://hub.docker.com/r/qwenllm/qwen3-asr). You only need to install the GPU driver and download the model files to run the code. Please follow the [NVIDIA Container Toolkit installation guide](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html) to ensure Docker can access your GPU. If you are in Mainland China and have trouble reaching Docker Hub, you may use a registry mirror to accelerate image pulls.
+
+First, pull the image and start a container:
+
+```bash
+LOCAL_WORKDIR=/path/to/your/workspace
+HOST_PORT=8000
+CONTAINER_PORT=80
+docker run --gpus all --name qwen3-asr \
+    -v /var/run/docker.sock:/var/run/docker.sock -p $HOST_PORT:$CONTAINER_PORT \
+    --mount type=bind,source=$LOCAL_WORKDIR,target=/data/shared/Qwen3-ASR \
+    --shm-size=4gb \
+    -it qwenllm/qwen3-asr:latest
+```
+
+After running the command, you will enter the container’s bash shell. Your local workspace (**replace** `/path/to/your/workspace` **with the actual path**) will be mounted inside the container at `/data/shared/Qwen3-ASR`. Port `8000` on the host is mapped to port `80` in the container, so you can access services running in the container via `http://<host-ip>:8000`. Note that services inside the container must bind to `0.0.0.0` (not `127.0.0.1`) for port forwarding to work.
+
+If you exit the container, you can start it again and re-enter it with:
+
+```bash
+docker start qwen3-asr
+docker exec -it qwen3-asr bash
+```
+
+To remove the container completely, run:
+
+```bash
+docker rm -f qwen3-asr
+```
+
+
+## Evaluation
+
+During evaluation, we ran inference for all models with `dtype=torch.bfloat16` and set `max_new_tokens=1024` using vLLM. Greedy search was used for all decoding, and none of the tests specified a language parameter. The detailed evaluation results are shown below.
+
+<details>
+<summary>ASR Benchmarks on Public Datasets (WER ↓)</summary>
+
+<table>
+  <thead>
+    <tr>
+      <th colspan="2" style="text-align: left;"></th>
+      <th style="text-align: center;">GPT-4o<br>-Transcribe</th>
+      <th style="text-align: center;">Gemini-2.5<br>-Pro</th>
+      <th style="text-align: center;">Doubao-ASR</th>
+      <th style="text-align: center;">Whisper<br>-large-v3</th>
+      <th style="text-align: center;">Fun-ASR<br>-MLT-Nano</th>
+      <th style="text-align: center;">Qwen3-ASR<br>-0.6B</th>
+      <th style="text-align: center;">Qwen3-ASR<br>-1.7B</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td colspan="9" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">English (en)</td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">Librispeech<br>clean | other</td>
+      <td style="text-align: center;"><strong>1.39</strong> | 3.75</td>
+      <td style="text-align: center;">2.89 | 3.56</td>
+      <td style="text-align: center;">2.78 | 5.70</td>
+      <td style="text-align: center;">1.51 | 3.97</td>
+      <td style="text-align: center;">1.68 | 4.03</td>
+      <td style="text-align: center;">2.11 | 4.55</td>
+      <td style="text-align: center;">1.63 | <strong>3.38</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">GigaSpeech</td>
+      <td style="text-align: center;">25.50</td>
+      <td style="text-align: center;">9.37</td>
+      <td style="text-align: center;">9.55</td>
+      <td style="text-align: center;">9.76</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">8.88</td>
+      <td style="text-align: center;"><strong>8.45</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">CV-en</td>
+      <td style="text-align: center;">9.08</td>
+      <td style="text-align: center;">14.49</td>
+      <td style="text-align: center;">13.78</td>
+      <td style="text-align: center;">9.90</td>
+      <td style="text-align: center;">9.90</td>
+      <td style="text-align: center;">9.92</td>
+      <td style="text-align: center;"><strong>7.39</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">Fleurs-en</td>
+      <td style="text-align: center;"><strong>2.40</strong></td>
+      <td style="text-align: center;">2.94</td>
+      <td style="text-align: center;">6.31</td>
+      <td style="text-align: center;">4.08</td>
+      <td style="text-align: center;">5.49</td>
+      <td style="text-align: center;">4.39</td>
+      <td style="text-align: center;">3.35</td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">MLS-en</td>
+      <td style="text-align: center;">5.12</td>
+      <td style="text-align: center;"><strong>3.68</strong></td>
+      <td style="text-align: center;">7.09</td>
+      <td style="text-align: center;">4.87</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">6.00</td>
+      <td style="text-align: center;">4.58</td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">Tedlium</td>
+      <td style="text-align: center;">7.69</td>
+      <td style="text-align: center;">6.15</td>
+      <td style="text-align: center;">4.91</td>
+      <td style="text-align: center;">6.84</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>3.85<strong></td>
+      <td style="text-align: center;"><strong>4.50</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">VoxPopuli</td>
+      <td style="text-align: center;">10.29</td>
+      <td style="text-align: center;">11.36</td>
+      <td style="text-align: center;">12.12</td>
+      <td style="text-align: center;">12.05</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>9.96<strong></td>
+      <td style="text-align: center;"><strong>9.15</strong></td>
+    </tr>
+    <tr>
+      <td colspan="9" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">Chinese (zh)</td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">WenetSpeech<br>net | meeting</td>
+      <td style="text-align: center;">15.30 | 32.27</td>
+      <td style="text-align: center;">14.43 | 13.47</td>
+      <td style="text-align: center;">N/A</td>
+      <td style="text-align: center;">9.86 | 19.11</td>
+      <td style="text-align: center;">6.35 | -</td>
+      <td style="text-align: center;">5.97 | 6.88</td>
+      <td style="text-align: center;"><strong>4.97</strong> | <strong>5.88</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">AISHELL-2-test</td>
+      <td style="text-align: center;">4.24</td>
+      <td style="text-align: center;">11.62</td>
+      <td style="text-align: center;">2.85</td>
+      <td style="text-align: center;">5.06</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">3.15</td>
+      <td style="text-align: center;"><strong>2.71</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">SpeechIO</td>
+      <td style="text-align: center;">12.86</td>
+      <td style="text-align: center;">5.30</td>
+      <td style="text-align: center;">2.93</td>
+      <td style="text-align: center;">7.56</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">3.44</td>
+      <td style="text-align: center;"><strong>2.88</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">Fleurs-zh</td>
+      <td style="text-align: center;">2.44</td>
+      <td style="text-align: center;">2.71</td>
+      <td style="text-align: center;">2.69</td>
+      <td style="text-align: center;">4.09</td>
+      <td style="text-align: center;">3.51</td>
+      <td style="text-align: center;">2.88</td>
+      <td style="text-align: center;"><strong>2.41</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">CV-zh</td>
+      <td style="text-align: center;">6.32</td>
+      <td style="text-align: center;">7.70</td>
+      <td style="text-align: center;">5.95</td>
+      <td style="text-align: center;">12.91</td>
+      <td style="text-align: center;">6.20</td>
+      <td style="text-align: center;">6.89</td>
+      <td style="text-align: center;"><strong>5.35</strong></td>
+    </tr>
+    <tr>
+      <td colspan="9" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">Chinese Dialect</td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">KeSpeech</td>
+      <td style="text-align: center;">26.87</td>
+      <td style="text-align: center;">24.71</td>
+      <td style="text-align: center;">5.27</td>
+      <td style="text-align: center;">28.79</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">7.08</td>
+      <td style="text-align: center;"><strong>5.10</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">Fleurs-yue</td>
+      <td style="text-align: center;">4.98</td>
+      <td style="text-align: center;">9.43</td>
+      <td style="text-align: center;">4.98</td>
+      <td style="text-align: center;">9.18</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">5.79</td>
+      <td style="text-align: center;"><strong>3.98</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">CV-yue</td>
+      <td style="text-align: center;">11.36</td>
+      <td style="text-align: center;">18.76</td>
+      <td style="text-align: center;">13.20</td>
+      <td style="text-align: center;">16.23</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">9.50</td>
+      <td style="text-align: center;"><strong>7.57</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">CV-zh-tw</td>
+      <td style="text-align: center;">6.32</td>
+      <td style="text-align: center;">7.31</td>
+      <td style="text-align: center;">4.06</td>
+      <td style="text-align: center;">7.84</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">5.59</td>
+      <td style="text-align: center;"><strong>3.77</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">WenetSpeech-Yue<br>short | long</td>
+      <td style="text-align: center;">15.62 | 25.29</td>
+      <td style="text-align: center;">25.19 | 11.23</td>
+      <td style="text-align: center;">9.74 | 11.40</td>
+      <td style="text-align: center;">32.26 | 46.64</td>
+      <td style="text-align: center;">- | -</td>
+      <td style="text-align: center;">7.54 | 9.92</td>
+      <td style="text-align: center;"><strong>5.82</strong> | <strong>8.85</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">WenetSpeech-Chuan<br>easy | hard</td>
+      <td style="text-align: center;">34.81 | 53.98</td>
+      <td style="text-align: center;">43.79 | 67.30</td>
+      <td style="text-align: center;"><strong>11.40<strong> | <strong>20.20</strong></td>
+      <td style="text-align: center;">14.35 | 26.80</td>
+      <td style="text-align: center;">- | -</td>
+      <td style="text-align: center;">13.92 | 24.45</td>
+      <td style="text-align: center;">11.99 | 21.63</td>
+    </tr>
+  </tbody>
+</table>
+
+</details>
+
+<details>
+<summary>ASR Benchmarks on Internal Datasets (WER ↓)</summary>
+
+<table>
+  <thead>
+    <tr>
+      <th style="text-align: left;"></th>
+      <th style="text-align: center;">GPT-4o<br>-Transcribe</th>
+      <th style="text-align: center;">Gemini-2.5<br>-Pro</th>
+      <th style="text-align: center;">Doubao-ASR</th>
+      <th style="text-align: center;">Whisper<br>-large-v3</th>
+      <th style="text-align: center;">Fun-ASR<br>-MLT-Nano</th>
+      <th style="text-align: center;">Qwen3-ASR<br>-0.6B</th>
+      <th style="text-align: center;">Qwen3-ASR<br>-1.7B</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td colspan="8" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">Accented English</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Dialog-Accented English</td>
+      <td style="text-align: center;">28.56</td>
+      <td style="text-align: center;">23.85</td>
+      <td style="text-align: center;">20.41</td>
+      <td style="text-align: center;">21.30</td>
+      <td style="text-align: center;">19.96</td>
+      <td style="text-align: center;"><strong>16.62<strong></td>
+      <td style="text-align: center;"><strong>16.07</strong></td>
+    </tr>
+    <tr>
+      <td colspan="8" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">Chinese Mandarin</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Elders&Kids</td>
+      <td style="text-align: center;">14.27</td>
+      <td style="text-align: center;">36.93</td>
+      <td style="text-align: center;">4.17</td>
+      <td style="text-align: center;">10.61</td>
+      <td style="text-align: center;">4.54</td>
+      <td style="text-align: center;">4.48</td>
+      <td style="text-align: center;"><strong>3.81</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">ExtremeNoise</td>
+      <td style="text-align: center;">36.11</td>
+      <td style="text-align: center;">29.06</td>
+      <td style="text-align: center;">17.04</td>
+      <td style="text-align: center;">63.17</td>
+      <td style="text-align: center;">36.55</td>
+      <td style="text-align: center;">17.88</td>
+      <td style="text-align: center;"><strong>16.17</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">TongueTwister</td>
+      <td style="text-align: center;">20.87</td>
+      <td style="text-align: center;">4.97</td>
+      <td style="text-align: center;">3.47</td>
+      <td style="text-align: center;">16.63</td>
+      <td style="text-align: center;">9.02</td>
+      <td style="text-align: center;">4.06</td>
+      <td style="text-align: center;"><strong>2.44</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Dialog-Mandarin</td>
+      <td style="text-align: center;">20.73</td>
+      <td style="text-align: center;">12.50</td>
+      <td style="text-align: center;">6.61</td>
+      <td style="text-align: center;">14.01</td>
+      <td style="text-align: center;">7.32</td>
+      <td style="text-align: center;">7.06</td>
+      <td style="text-align: center;"><strong>6.54</strong></td>
+    </tr>
+    <tr>
+      <td colspan="8" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">Chinese Dialect</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Dialog-Cantonese</td>
+      <td style="text-align: center;">16.05</td>
+      <td style="text-align: center;">14.98</td>
+      <td style="text-align: center;">7.56</td>
+      <td style="text-align: center;">31.04</td>
+      <td style="text-align: center;">5.85</td>
+      <td style="text-align: center;"><strong>4.80<strong></td>
+      <td style="text-align: center;"><strong>4.12</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Dialog-Chinese Dialects</td>
+      <td style="text-align: center;">45.37</td>
+      <td style="text-align: center;">47.70</td>
+      <td style="text-align: center;">19.85</td>
+      <td style="text-align: center;">44.55</td>
+      <td style="text-align: center;">19.41</td>
+      <td style="text-align: center;"><strong>18.24<strong></td>
+      <td style="text-align: center;"><strong>15.94</strong></td>
+    </tr>
+  </tbody>
+</table>
+<p><strong>Dialect coverage:</strong> Results for <em>Dialog-Accented English</em> are averaged over 16 accents, and results for <em>Dialog-Chinese Dialects</em> are averaged over 22 Chinese dialects.</p>
+
+</details>
+
+<details>
+<summary>Multilingual ASR Benchmarks (WER ↓)</summary>
+
+<table>
+  <thead>
+    <tr>
+      <th style="text-align: left;"></th>
+      <th style="text-align: center;">GLM-ASR<br>-Nano-2512</th>
+      <th style="text-align: center;">Whisper<br>-large-v3</th>
+      <th style="text-align: center;">Fun-ASR<br>-MLT-Nano</th>
+      <th style="text-align: center;">Qwen3-ASR<br>-0.6B</th>
+      <th style="text-align: center;">Qwen3-ASR<br>-1.7B</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td colspan="6" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">Open-sourced Benchmarks</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">MLS</td>
+      <td style="text-align: center;">13.32</td>
+      <td style="text-align: center;">8.62</td>
+      <td style="text-align: center;">28.70</td>
+      <td style="text-align: center;">13.19</td>
+      <td style="text-align: center;"><strong>8.55</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">CommonVoice</td>
+      <td style="text-align: center;">19.40</td>
+      <td style="text-align: center;">10.77</td>
+      <td style="text-align: center;">17.25</td>
+      <td style="text-align: center;">12.75</td>
+      <td style="text-align: center;"><strong>9.18</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">MLC-SLM</td>
+      <td style="text-align: center;">34.93</td>
+      <td style="text-align: center;">15.68</td>
+      <td style="text-align: center;">29.94</td>
+      <td style="text-align: center;">15.84</td>
+      <td style="text-align: center;"><strong>12.74</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Fleurs</td>
+      <td style="text-align: center;">16.08</td>
+      <td style="text-align: center;">5.27</td>
+      <td style="text-align: center;">10.03</td>
+      <td style="text-align: center;">7.57</td>
+      <td style="text-align: center;"><strong>4.90</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Fleurs<sup>†</sup></td>
+      <td style="text-align: center;">20.05</td>
+      <td style="text-align: center;">6.85</td>
+      <td style="text-align: center;">31.89</td>
+      <td style="text-align: center;">10.37</td>
+      <td style="text-align: center;"><strong>6.62</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Fleurs<sup>††</sup></td>
+      <td style="text-align: center;">24.83</td>
+      <td style="text-align: center;"><strong>8.16</strong></td>
+      <td style="text-align: center;">47.84</td>
+      <td style="text-align: center;">21.80</td>
+      <td style="text-align: center;">12.60</td>
+    </tr>
+    <tr>
+      <td colspan="6" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">Qwen-ASR Internal Benchmarks</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">News-Multilingual</td>
+      <td style="text-align: center;">49.40</td>
+      <td style="text-align: center;">14.80</td>
+      <td style="text-align: center;">65.07</td>
+      <td style="text-align: center;">17.39</td>
+      <td style="text-align: center;"><strong>12.80</strong></td>
+    </tr>
+  </tbody>
+</table>
+<p><strong>Language coverage:</strong> <em>MLS</em> includes 8 languages: {da, de, en, es, fr, it, pl, pt}.<br><em>CommonVoice</em> includes 13 languages: {en, zh, yue, zh_TW, ar, de, es, fr, it, ja, ko, pt, ru}.<br><em>MLC-SLM</em> includes 11 languages: {en, fr, de, it, pt, es, ja, ko, ru, th, vi}.<br><em>Fleurs</em> includes 12 languages: {en, zh, yue, ar, de, es, fr, it, ja, ko, pt, ru }.<br><em>Fleurs<sup>†</sup></em> includes 8 additional languages beyond Fleurs: {hi, id, ms, nl, pl, th, tr, vi}.<br><em>Fleurs<sup>††</sup></em> includes 10 additional languages beyond Fleurs<sup>†</sup>: {cs, da, el, fa, fi, fil, hu, mk, ro, sv}.<br><em>News-Multilingual</em> includes 15 languages: {ar, de, es, fr, hi, id, it, ja, ko, nl, pl, pt, ru, th, vi}.</p>
+
+</details>
+
+<details>
+<summary>Language Identification Accuracy (%) ↑</summary>
+
+<table>
+  <thead>
+    <tr>
+      <th style="text-align: left;"></th>
+      <th style="text-align: center;">Whisper-large-v3</th>
+      <th style="text-align: center;">Qwen3-ASR-0.6B</th>
+      <th style="text-align: center;">Qwen3-ASR-1.7B</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td style="text-align: left;">MLS</td>
+      <td style="text-align: center;"><strong>99.9</strong></td>
+      <td style="text-align: center;">99.3</td>
+      <td style="text-align: center;"><strong>99.9</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">CommonVoice</td>
+      <td style="text-align: center;">92.7</td>
+      <td style="text-align: center;"><strong>98.2<strong></td>
+      <td style="text-align: center;"><strong>98.7</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">MLC-SLM</td>
+      <td style="text-align: center;">89.2</td>
+      <td style="text-align: center;"><strong>92.7<strong></td>
+      <td style="text-align: center;"><strong>94.1</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Fleurs</td>
+      <td style="text-align: center;">94.6</td>
+      <td style="text-align: center;"><strong>97.1<strong></td>
+      <td style="text-align: center;"><strong>98.7</strong></td>
+    </tr>
+    <tr style="border-top: 1px solid #ddd;">
+      <td style="text-align: left;"><em>Avg.</em></td>
+      <td style="text-align: center;">94.1</td>
+      <td style="text-align: center;"><strong>96.8<strong></td>
+      <td style="text-align: center;"><strong>97.9</strong></td>
+    </tr>
+  </tbody>
+</table>
+<p><strong>Language coverage:</strong> The language sets follow Multilingual ASR Benchmarks. Here, Fleurs corresponds to Fleurs<sup>††</sup> in Multilingual ASR Benchmarks and covers 30 languages.</p>
+
+</details>
+
+<details>
+<summary>Singing Voice & Song Transcription (WER ↓)</summary>
+
+<table>
+  <thead>
+    <tr>
+      <th style="text-align: left;"></th>
+      <th style="text-align: center;">GPT-4o<br>-Transcribe</th>
+      <th style="text-align: center;">Gemini-2.5<br>-Pro</th>
+      <th style="text-align: center;">Doubao-ASR<br>-1.0</th>
+      <th style="text-align: center;">Whisper<br>-large-v3</th>
+      <th style="text-align: center;">Fun-ASR-MLT<br>-Nano</th>
+      <th style="text-align: center;">Qwen3-ASR<br>-1.7B</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td colspan="7" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">Singing</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">M4Singer</td>
+      <td style="text-align: center;">16.77</td>
+      <td style="text-align: center;">20.88</td>
+      <td style="text-align: center;">7.88</td>
+      <td style="text-align: center;">13.58</td>
+      <td style="text-align: center;">7.29</td>
+      <td style="text-align: center;"><strong>5.98</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">MIR-1k-vocal</td>
+      <td style="text-align: center;">11.87</td>
+      <td style="text-align: center;">9.85</td>
+      <td style="text-align: center;">6.56</td>
+      <td style="text-align: center;">11.71</td>
+      <td style="text-align: center;">8.17</td>
+      <td style="text-align: center;"><strong>6.25</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Opencpop</td>
+      <td style="text-align: center;">7.93</td>
+      <td style="text-align: center;">6.49</td>
+      <td style="text-align: center;">3.80</td>
+      <td style="text-align: center;">9.52</td>
+      <td style="text-align: center;"><strong>2.98</strong></td>
+      <td style="text-align: center;">3.08</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Popcs</td>
+      <td style="text-align: center;">32.84</td>
+      <td style="text-align: center;">15.13</td>
+      <td style="text-align: center;">8.97</td>
+      <td style="text-align: center;">13.77</td>
+      <td style="text-align: center;">9.42</td>
+      <td style="text-align: center;"><strong>8.52</strong></td>
+    </tr>
+    <tr>
+      <td colspan="7" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">Songs with BGM</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">EntireSongs-en</td>
+      <td style="text-align: center;">30.71</td>
+      <td style="text-align: center;"><strong>12.18</strong></td>
+      <td style="text-align: center;">33.51</td>
+      <td style="text-align: center;">N/A</td>
+      <td style="text-align: center;">N/A</td>
+      <td style="text-align: center;">14.60</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">EntireSongs-zh</td>
+      <td style="text-align: center;">34.86</td>
+      <td style="text-align: center;">18.68</td>
+      <td style="text-align: center;">23.99</td>
+      <td style="text-align: center;">N/A</td>
+      <td style="text-align: center;">N/A</td>
+      <td style="text-align: center;"><strong>13.91</strong></td>
+    </tr>
+  </tbody>
+</table>
+
+</details>
+
+<details>
+<summary>ASR Inference Mode Performance (WER ↓)</summary>
+
+<table>
+  <thead>
+    <tr>
+      <th style="text-align: left;">Model</th>
+      <th style="text-align: left;">Infer. Mode</th>
+      <th style="text-align: center;">Librispeech</th>
+      <th style="text-align: center;">Fleurs-en</th>
+      <th style="text-align: center;">Fleurs-zh</th>
+      <th style="text-align: center;">Avg.</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td rowspan="2" style="text-align: left; vertical-align: middle;">Qwen3-ASR-1.7B</td>
+      <td style="text-align: left;">Offline</td>
+      <td style="text-align: center;">1.63 | 3.38</td>
+      <td style="text-align: center;">3.35</td>
+      <td style="text-align: center;">2.41</td>
+      <td style="text-align: center;">2.69</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Streaming</td>
+      <td style="text-align: center;">1.95 | 4.51</td>
+      <td style="text-align: center;">4.02</td>
+      <td style="text-align: center;">2.84</td>
+      <td style="text-align: center;">3.33</td>
+    </tr>
+    <tr style="border-top: 1px solid #ddd;">
+      <td rowspan="2" style="text-align: left; vertical-align: middle;">Qwen3-ASR-0.6B</td>
+      <td style="text-align: left;">Offline</td>
+      <td style="text-align: center;">2.11 | 4.55</td>
+      <td style="text-align: center;">4.39</td>
+      <td style="text-align: center;">2.88</td>
+      <td style="text-align: center;">3.48</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Streaming</td>
+      <td style="text-align: center;">2.54 | 6.27</td>
+      <td style="text-align: center;">5.38</td>
+      <td style="text-align: center;">3.40</td>
+      <td style="text-align: center;">4.40</td>
+    </tr>
+  </tbody>
+</table>
+
+</details>
+
+<details>
+<summary>Forced Alignment Benchmarks (AAS ms ↓)</summary>
+
+<table>
+  <thead>
+    <tr>
+      <th style="text-align: left;"></th>
+      <th style="text-align: center;">Monotonic-Aligner</th>
+      <th style="text-align: center;">NFA</th>
+      <th style="text-align: center;">WhisperX</th>
+      <th style="text-align: center;">Qwen3-ForcedAligner-0.6B</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td colspan="5" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">MFA-Labeled Raw</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Chinese</td>
+      <td style="text-align: center;">161.1</td>
+      <td style="text-align: center;">109.8</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>33.1</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">English</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">107.5</td>
+      <td style="text-align: center;">92.1</td>
+      <td style="text-align: center;"><strong>37.5</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">French</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">100.7</td>
+      <td style="text-align: center;">145.3</td>
+      <td style="text-align: center;"><strong>41.7</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">German</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">122.7</td>
+      <td style="text-align: center;">165.1</td>
+      <td style="text-align: center;"><strong>46.5</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Italian</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">142.7</td>
+      <td style="text-align: center;">155.5</td>
+      <td style="text-align: center;"><strong>75.5</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Japanese</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>42.2</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Korean</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>37.2</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Portuguese</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>38.4</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Russian</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">200.7</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>40.2</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Spanish</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">124.7</td>
+      <td style="text-align: center;">108.0</td>
+      <td style="text-align: center;"><strong>36.8</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;"><em>Avg.</em></td>
+      <td style="text-align: center;">161.1</td>
+      <td style="text-align: center;">129.8</td>
+      <td style="text-align: center;">133.2</td>
+      <td style="text-align: center;"><strong>42.9</strong></td>
+    </tr>
+    <tr>
+      <td colspan="5" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">MFA-Labeled Concat-300s</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Chinese</td>
+      <td style="text-align: center;">1742.4</td>
+      <td style="text-align: center;">235.0</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>36.5</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">English</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">226.7</td>
+      <td style="text-align: center;">227.2</td>
+      <td style="text-align: center;"><strong>58.6</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">French</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">230.6</td>
+      <td style="text-align: center;">2052.2</td>
+      <td style="text-align: center;"><strong>53.4</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">German</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">220.3</td>
+      <td style="text-align: center;">993.4</td>
+      <td style="text-align: center;"><strong>62.4</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Italian</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">290.5</td>
+      <td style="text-align: center;">5719.4</td>
+      <td style="text-align: center;"><strong>81.6</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Japanese</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>81.3</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Korean</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>42.2</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Portuguese</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>50.0</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Russian</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">283.3</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>43.0</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Spanish</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">240.2</td>
+      <td style="text-align: center;">4549.9</td>
+      <td style="text-align: center;"><strong>39.6</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Cross-lingual</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>34.2</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;"><em>Avg.</em></td>
+      <td style="text-align: center;">1742.4</td>
+      <td style="text-align: center;">246.7</td>
+      <td style="text-align: center;">2708.4</td>
+      <td style="text-align: center;"><strong>52.9</strong></td>
+    </tr>
+    <tr>
+      <td colspan="5" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">Human-Labeled</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Raw</td>
+      <td style="text-align: center;">49.9</td>
+      <td style="text-align: center;">88.6</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>27.8</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Raw-Noisy</td>
+      <td style="text-align: center;">53.3</td>
+      <td style="text-align: center;">89.5</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>41.8</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Concat-60s</td>
+      <td style="text-align: center;">51.1</td>
+      <td style="text-align: center;">86.7</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>25.3</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Concat-300s</td>
+      <td style="text-align: center;">410.8</td>
+      <td style="text-align: center;">140.0</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>24.8</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Concat-Cross-lingual</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>42.5</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;"><em>Avg.</em></td>
+      <td style="text-align: center;">141.3</td>
+      <td style="text-align: center;">101.2</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>32.4</strong></td>
+    </tr>
+  </tbody>
+</table>
+
+</details>
+
+
+## Citation
+
+If you find our paper and code useful in your research, please consider giving a star :star: and citation :pencil: :)
+
+```BibTeX
+@article{Qwen3-ASR,
+  title={Qwen3-ASR Technical Report},
+  author={Xian Shi, Xiong Wang, Zhifang Guo, Yongqi Wang, Pei Zhang, Xinyu Zhang, Zishan Guo, Hongkun Hao, Yu Xi, Baosong Yang, Jin Xu, Jingren Zhou, Junyang Lin},
+  journal={arXiv preprint arXiv:2601.21337},
+  year={2026}
+}
+```
+
+
+<br>

qwen-aligner/chat_template.json

@@ -0,0 +1 @@
+{"chat_template": "{%- set ns = namespace(system_text=\"\") -%}\n{%- for m in messages -%}\n  {%- if m.role == 'system' -%}\n    {%- if m.content is string -%}\n      {%- set ns.system_text = ns.system_text + m.content -%}\n    {%- else -%}\n      {%- for c in m.content -%}\n        {%- if c.type == 'text' and (c.text is defined) -%}\n          {%- set ns.system_text = ns.system_text + c.text -%}\n        {%- endif -%}\n      {%- endfor -%}\n    {%- endif -%}\n  {%- endif -%}\n{%- endfor -%}\n\n{%- set ns2 = namespace(audio_tokens=\"\") -%}\n{%- for m in messages -%}\n  {%- if m.content is not string -%}\n    {%- for c in m.content -%}\n      {%- if c.type == 'audio' or ('audio' in c) or ('audio_url' in c) -%}\n        {%- set ns2.audio_tokens = ns2.audio_tokens + \"<|audio_start|><|audio_pad|><|audio_end|>\" -%}\n      {%- endif -%}\n    {%- endfor -%}\n  {%- endif -%}\n{%- endfor -%}\n\n{{- '<|im_start|>system\\n' + (ns.system_text if ns.system_text is string else '') + '<|im_end|>\\n' -}}\n{{- '<|im_start|>user\\n' + ns2.audio_tokens + '<|im_end|>\\n' -}}\n{%- if add_generation_prompt -%}\n{{- '<|im_start|>assistant\\n' -}}\n{%- endif -%}"}

qwen-aligner/config.json

@@ -0,0 +1,205 @@
+{
+    "architectures": [
+      "Qwen3ASRForConditionalGeneration"
+    ],
+    "timestamp_token_id": 151705,
+    "timestamp_segment_time": 80,
+    "model_type": "qwen3_asr",
+    "support_languages": [
+      "Chinese",
+      "Cantonese",
+      "English",
+      "German",
+      "Spanish",
+      "French",
+      "Italian",
+      "Portuguese",
+      "Russian",
+      "Korean",
+      "Japanese"
+    ],
+    "thinker_config": {
+      "model_type": "qwen3_forced_aligner",
+      "classify_num": 5000,
+      "architectures": [
+        "Qwen3ASRForConditionalGeneration"
+      ],
+      "audio_config": {
+        "_name_or_path": "",
+        "activation_dropout": 0,
+        "activation_function": "gelu",
+        "add_cross_attention": false,
+        "architectures": null,
+        "attention_dropout": 0,
+        "bad_words_ids": null,
+        "begin_suppress_tokens": null,
+        "bos_token_id": null,
+        "chunk_size_feed_forward": 0,
+        "conv_chunksize": 500,
+        "cross_attention_hidden_size": null,
+        "d_model": 1024,
+        "decoder_start_token_id": null,
+        "diversity_penalty": 0.0,
+        "do_sample": false,
+        "downsample_hidden_size": 480,
+        "dropout": 0,
+        "dtype": null,
+        "early_stopping": false,
+        "encoder_attention_heads": 16,
+        "encoder_ffn_dim": 4096,
+        "encoder_layers": 24,
+        "encoder_no_repeat_ngram_size": 0,
+        "eos_token_id": null,
+        "exponential_decay_length_penalty": null,
+        "finetuning_task": null,
+        "forced_bos_token_id": null,
+        "forced_eos_token_id": null,
+        "id2label": {
+          "0": "LABEL_0",
+          "1": "LABEL_1"
+        },
+        "initializer_range": 0.02,
+        "is_decoder": false,
+        "is_encoder_decoder": false,
+        "label2id": {
+          "LABEL_0": 0,
+          "LABEL_1": 1
+        },
+        "length_penalty": 1.0,
+        "max_length": 20,
+        "max_source_positions": 1500,
+        "min_length": 0,
+        "model_type": "qwen3_asr_audio_encoder",
+        "n_window": 50,
+        "n_window_infer": 800,
+        "no_repeat_ngram_size": 0,
+        "num_beam_groups": 1,
+        "num_beams": 1,
+        "num_hidden_layers": 24,
+        "num_mel_bins": 128,
+        "num_return_sequences": 1,
+        "output_attentions": false,
+        "output_dim": 1024,
+        "output_hidden_states": false,
+        "output_scores": false,
+        "pad_token_id": null,
+        "prefix": null,
+        "problem_type": null,
+        "pruned_heads": {},
+        "remove_invalid_values": false,
+        "repetition_penalty": 1.0,
+        "return_dict": true,
+        "return_dict_in_generate": false,
+        "scale_embedding": false,
+        "sep_token_id": null,
+        "suppress_tokens": null,
+        "task_specific_params": null,
+        "temperature": 1.0,
+        "tf_legacy_loss": false,
+        "tie_encoder_decoder": false,
+        "tie_word_embeddings": false,
+        "tokenizer_class": null,
+        "top_k": 50,
+        "top_p": 1.0,
+        "torchscript": false,
+        "typical_p": 1.0,
+        "use_bfloat16": false
+      },
+      "audio_end_token_id": 151670,
+      "audio_start_token_id": 151669,
+      "audio_token_id": 151676,
+      "dtype": "bfloat16",
+      "initializer_range": 0.02,
+      "text_config": {
+        "_name_or_path": "",
+        "add_cross_attention": false,
+        "architectures": null,
+        "attention_bias": false,
+        "attention_dropout": 0.0,
+        "bad_words_ids": null,
+        "begin_suppress_tokens": null,
+        "bos_token_id": null,
+        "chunk_size_feed_forward": 0,
+        "cross_attention_hidden_size": null,
+        "decoder_start_token_id": null,
+        "diversity_penalty": 0.0,
+        "do_sample": false,
+        "dtype": null,
+        "early_stopping": false,
+        "encoder_no_repeat_ngram_size": 0,
+        "eos_token_id": null,
+        "exponential_decay_length_penalty": null,
+        "finetuning_task": null,
+        "forced_bos_token_id": null,
+        "forced_eos_token_id": null,
+        "head_dim": 128,
+        "hidden_act": "silu",
+        "hidden_size": 1024,
+        "id2label": {
+          "0": "LABEL_0",
+          "1": "LABEL_1"
+        },
+        "initializer_range": 0.02,
+        "intermediate_size": 3072,
+        "is_decoder": false,
+        "is_encoder_decoder": false,
+        "label2id": {
+          "LABEL_0": 0,
+          "LABEL_1": 1
+        },
+        "length_penalty": 1.0,
+        "max_length": 20,
+        "max_position_embeddings": 8192,
+        "min_length": 0,
+        "model_type": "qwen3",
+        "no_repeat_ngram_size": 0,
+        "num_attention_heads": 16,
+        "num_beam_groups": 1,
+        "num_beams": 1,
+        "num_hidden_layers": 28,
+        "num_key_value_heads": 8,
+        "num_return_sequences": 1,
+        "output_attentions": false,
+        "output_hidden_states": false,
+        "output_scores": false,
+        "pad_token_id": null,
+        "prefix": null,
+        "problem_type": null,
+        "pruned_heads": {},
+        "remove_invalid_values": false,
+        "repetition_penalty": 1.0,
+        "return_dict": true,
+        "return_dict_in_generate": false,
+        "rms_norm_eps": 1e-06,
+        "rope_scaling": {
+          "interleaved": true,
+          "mrope_interleaved": true,
+          "mrope_section": [
+            24,
+            20,
+            20
+          ],
+          "rope_type": "default",
+          "type": "default"
+        },
+        "rope_theta": 1000000,
+        "sep_token_id": null,
+        "suppress_tokens": null,
+        "task_specific_params": null,
+        "temperature": 1.0,
+        "tf_legacy_loss": false,
+        "tie_encoder_decoder": false,
+        "tie_word_embeddings": false,
+        "tokenizer_class": null,
+        "top_k": 50,
+        "top_p": 1.0,
+        "torchscript": false,
+        "typical_p": 1.0,
+        "use_bfloat16": false,
+        "use_cache": true,
+        "vocab_size": 152064
+      }
+    },
+    "transformers_version": "4.57.6"
+  }
+  

qwen-aligner/generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "eos_token_id": [151643,151645],
+  "pad_token_id": 151643,
+  "do_sample": false
+}

qwen-aligner/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:47831d0e82f96b20e9034dba01a075ee06436654719f6a68289e49f1b65ce0e7
+size 1835544544

qwen-aligner/preprocessor_config.json

@@ -0,0 +1,14 @@
+{
+  "chunk_length": 30,
+  "dither": 0.0,
+  "feature_extractor_type": "WhisperFeatureExtractor",
+  "feature_size": 128,
+  "hop_length": 160,
+  "n_fft": 400,
+  "n_samples": 480000,
+  "nb_max_frames": 3000,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "processor_class": "Qwen3ASRProcessor",
+  "return_attention_mask": true
+}

qwen-aligner/tokenizer_config.json

@@ -0,0 +1,557 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151665": {
+      "content": "<tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151666": {
+      "content": "</tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151667": {
+      "content": "<think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151668": {
+      "content": "</think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151669": {
+      "content": "<|audio_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151670": {
+      "content": "<|audio_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151671": {
+      "content": "<tts_pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151672": {
+      "content": "<tts_text_bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151673": {
+      "content": "<tts_text_eod>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151674": {
+      "content": "<tts_text_bos_single>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151675": {
+      "content": "<non_speech>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151676": {
+      "content": "<|audio_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151677": {
+      "content": "<blank1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151678": {
+      "content": "<blank2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151679": {
+      "content": "<blank3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151680": {
+      "content": "<blank4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151681": {
+      "content": "<blank5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151682": {
+      "content": "<blank6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151683": {
+      "content": "<blank7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151684": {
+      "content": "<blank8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151685": {
+      "content": "<blank9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151686": {
+      "content": "<blank10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151687": {
+      "content": "<blank11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151688": {
+      "content": "<blank12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151689": {
+      "content": "<blank13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151690": {
+      "content": "<blank14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151691": {
+      "content": "<blank15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151692": {
+      "content": "<blank16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151693": {
+      "content": "<blank17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151694": {
+      "content": "<blank18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151695": {
+      "content": "<blank19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151696": {
+      "content": "<blank20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151697": {
+      "content": "<blank21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151698": {
+      "content": "<blank22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151699": {
+      "content": "<blank23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151700": {
+      "content": "<blank24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151701": {
+      "content": "<blank25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151702": {
+      "content": "<blank26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151703": {
+      "content": "<blank27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151704": {
+      "content": "<asr_text>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151705": {
+      "content": "<timestamp>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>",
+    "<|audio_start|>",
+    "<|audio_end|>",
+    "<tts_pad>",
+    "<tts_text_bos>",
+    "<tts_text_bos_single>",
+    "<|audio_pad|>"
+  ],
+  "audio_bos_token": "<|audio_start|>",
+  "audio_eos_token": "<|audio_end|>",
+  "audio_token": "<|audio_pad|>",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {
+    "audio_bos_token": "<|audio_start|>",
+    "audio_eos_token": "<|audio_end|>",
+    "audio_token": "<|audio_pad|>",
+    "image_token": "<|image_pad|>",
+    "video_token": "<|video_pad|>",
+    "vision_bos_token": "<|vision_start|>",
+    "vision_eos_token": "<|vision_end|>"
+  },
+  "image_token": "<|image_pad|>",
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "processor_class": "Qwen3ASRProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null,
+  "video_token": "<|video_pad|>",
+  "vision_bos_token": "<|vision_start|>",
+  "vision_eos_token": "<|vision_end|>"
+}

qwen-asr/LICENSE

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

qwen-asr/PROVENANCE.txt

@@ -0,0 +1,4 @@
+Original model: Qwen/Qwen3-ASR-1.7B
+Upstream revision: 7278e1e70fe206f11671096ffdd38061171dd6e5
+License: apache-2.0
+Distributed for SnapTop without changing the upstream license.

qwen-asr/README.md

@@ -0,0 +1,1393 @@
+---
+license: apache-2.0
+pipeline_tag: automatic-speech-recognition
+---
+
+# Qwen3-ASR
+
+## Overview
+
+### Introduction
+
+<p align="center">
+    <img src="https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-ASR-Repo/qwen3_asr_introduction.png" width="90%"/>
+<p>
+
+The Qwen3-ASR family includes Qwen3-ASR-1.7B and Qwen3-ASR-0.6B, which support language identification and ASR for 52 languages and dialects. Both leverage large-scale speech training data and the strong audio understanding capability of their foundation model, Qwen3-Omni. Experiments show that the 1.7B version achieves state-of-the-art performance among open-source ASR models and is competitive with the strongest proprietary commercial APIs. Here are the main features:
+
+* **All-in-one**: Qwen3-ASR-1.7B and Qwen3-ASR-0.6B support language identification and speech recognition for 30 languages and 22 Chinese dialects, so as to English accents from multiple countries and regions.
+
+* **Excellent and Fast**: The Qwen3-ASR family ASR models maintains high-quality and robust recognition under complex acoustic environments and challenging text patterns. Qwen3-ASR-1.7B achieves strong performance on both open-sourced and internal benchmarks. While the 0.6B version achieves accuracy-efficient trade-off, it reaches 2000 times throughput at a concurrency of 128. They both achieve streaming / offline unified inference with single model and support transcribe long audio.
+
+* **Novel and strong forced alignment Solution**: We introduce Qwen3-ForcedAligner-0.6B, which supports timestamp prediction for arbitrary units within up to 5 minutes of speech in 11 languages. Evaluations show its timestamp accuracy surpasses E2E based forced-alignment models.
+
+* **Comprehensive inference toolkit**: In addition to open-sourcing the architectures and weights of the Qwen3-ASR series, we also release a powerful, full-featured inference framework that supports vLLM-based batch inference, asynchronous serving, streaming inference, timestamp prediction, and more.
+
+### Model Architecture
+
+<p align="center">
+    <img src="https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-ASR-Repo/overview.jpg" width="100%"/>
+<p>
+
+
+### Released Models Description and Download
+
+Below is an introduction and download information for the Qwen3-ASR models. Please select and download the model that fits your needs.
+
+| Model | Supported Languages | Supported Dialects | Inference Mode | Audio Types |
+|---|---|---|---|---|
+| Qwen3-ASR-1.7B & Qwen3-ASR-0.6B | Chinese (zh), English (en), Cantonese (yue), Arabic (ar), German (de), French (fr), Spanish (es), Portuguese (pt), Indonesian (id), Italian (it), Korean (ko), Russian (ru), Thai (th), Vietnamese (vi), Japanese (ja), Turkish (tr), Hindi (hi), Malay (ms), Dutch (nl), Swedish (sv), Danish (da), Finnish (fi), Polish (pl), Czech (cs), Filipino (fil), Persian (fa), Greek (el), Hungarian (hu), Macedonian (mk), Romanian (ro) | Anhui, Dongbei, Fujian, Gansu, Guizhou, Hebei, Henan, Hubei, Hunan, Jiangxi, Ningxia, Shandong, Shaanxi, Shanxi, Sichuan, Tianjin, Yunnan, Zhejiang, Cantonese (Hong Kong accent), Cantonese (Guangdong accent), Wu language, Minnan language. | Offline / Streaming | Speech, Singing Voice, Songs with BGM |
+| Qwen3-ForcedAligner-0.6B | Chinese, English, Cantonese, French, German, Italian, Japanese, Korean, Portuguese, Russian, Spanish | -- | NAR | Speech |
+
+During model loading in the `qwen-asr` package or vLLM, model weights will be downloaded automatically based on the model name. However, if your runtime environment does not allow downloading weights during execution, you can use the following commands to manually download the model weights to a local directory:
+
+```bash
+# Download through ModelScope (recommended for users in Mainland China)
+pip install -U modelscope
+modelscope download --model Qwen/Qwen3-ASR-1.7B  --local_dir ./Qwen3-ASR-1.7B
+modelscope download --model Qwen/Qwen3-ASR-0.6B --local_dir ./Qwen3-ASR-0.6B
+modelscope download --model Qwen/Qwen3-ForcedAligner-0.6B --local_dir ./Qwen3-ForcedAligner-0.6B
+# Download through Hugging Face
+pip install -U "huggingface_hub[cli]"
+huggingface-cli download Qwen/Qwen3-ASR-1.7B --local-dir ./Qwen3-ASR-1.7B
+huggingface-cli download Qwen/Qwen3-ASR-0.6B --local-dir ./Qwen3-ASR-0.6B
+huggingface-cli download Qwen/Qwen3-ForcedAligner-0.6B --local-dir ./Qwen3-ForcedAligner-0.6B
+```
+
+
+## Quickstart
+
+### Environment Setup
+
+The easiest way to use Qwen3-ASR is to install the `qwen-asr` Python package from PyPI. This will pull in the required runtime dependencies and allow you to load any released Qwen3-ASR model. If you’d like to simplify environment setup further, you can also use our official [Docker image](#docker). The `qwen-asr` package provides two backends: the transformers backend and the vLLM backend. For usage instructions for different backends, please refer to [Python Package Usage](#python-package-usage). We recommend using a **fresh, isolated environment** to avoid dependency conflicts with existing packages. You can create a clean Python 3.12 environment like this:
+
+```bash
+conda create -n qwen3-asr python=3.12 -y
+conda activate qwen3-asr
+```
+
+Run the following command to get the minimal installation with transformers-backend support:
+
+```bash
+pip install -U qwen-asr
+```
+
+To enable the vLLM backend for faster inference and streaming support, run:
+
+```bash
+pip install -U qwen-asr[vllm]
+```
+
+If you want to develop or modify the code locally, install from source in editable mode:
+
+```bash
+git clone https://github.com/QwenLM/Qwen3-ASR.git
+cd Qwen3-ASR
+pip install -e .
+# support vLLM backend
+# pip install -e ".[vllm]"
+```
+
+Additionally, we recommend using FlashAttention 2 to reduce GPU memory usage and accelerate inference speed, especially for long inputs and large batch sizes.
+
+```bash
+pip install -U flash-attn --no-build-isolation
+```
+
+If your machine has less than 96GB of RAM and lots of CPU cores, run:
+
+```bash
+MAX_JOBS=4 pip install -U flash-attn --no-build-isolation
+```
+
+Also, you should have hardware that is compatible with FlashAttention 2. Read more about it in the official documentation of the [FlashAttention repository](https://github.com/Dao-AILab/flash-attention). FlashAttention 2 can only be used when a model is loaded in `torch.float16` or `torch.bfloat16`.
+
+### Python Package Usage
+
+#### Quick Inference
+
+The `qwen-asr` package provides two backends: **transformers backend** and **vLLM backend**. You can pass audio inputs as a local path, a URL, base64 data, or a `(np.ndarray, sr)` tuple, and run batch inference. To quickly try Qwen3-ASR, you can use `Qwen3ASRModel.from_pretrained(...)` for the transformers backend with the following code:
+
+```python
+import torch
+from qwen_asr import Qwen3ASRModel
+
+model = Qwen3ASRModel.from_pretrained(
+    "Qwen/Qwen3-ASR-1.7B",
+    dtype=torch.bfloat16,
+    device_map="cuda:0",
+    # attn_implementation="flash_attention_2",
+    max_inference_batch_size=32, # Batch size limit for inference. -1 means unlimited. Smaller values can help avoid OOM.
+    max_new_tokens=256, # Maximum number of tokens to generate. Set a larger value for long audio input.
+)
+
+results = model.transcribe(
+    audio="https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-ASR-Repo/asr_en.wav",
+    language=None, # set "English" to force the language
+)
+
+print(results[0].language)
+print(results[0].text)
+```
+
+If you want to return timestamps, pass `forced_aligner` and its init kwargs. Here is an example of batch inference with timestamps output:
+
+```python
+import torch
+from qwen_asr import Qwen3ASRModel
+
+model = Qwen3ASRModel.from_pretrained(
+    "Qwen/Qwen3-ASR-1.7B",
+    dtype=torch.bfloat16,
+    device_map="cuda:0",
+    # attn_implementation="flash_attention_2",
+    max_inference_batch_size=32, # Batch size limit for inference. -1 means unlimited. Smaller values can help avoid OOM.
+    max_new_tokens=256, # Maximum number of tokens to generate. Set a larger value for long audio input.
+    forced_aligner="Qwen/Qwen3-ForcedAligner-0.6B",
+    forced_aligner_kwargs=dict(
+        dtype=torch.bfloat16,
+        device_map="cuda:0",
+        # attn_implementation="flash_attention_2",
+    ),
+)
+
+results = model.transcribe(
+    audio=[
+      "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-ASR-Repo/asr_zh.wav",
+      "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-ASR-Repo/asr_en.wav",
+    ],
+    language=["Chinese", "English"], # can also be set to None for automatic language detection
+    return_time_stamps=True,
+)
+
+for r in results:
+    print(r.language, r.text, r.time_stamps[0])
+```
+
+For more detailed usage examples, please refer to the [example code](https://github.com/QwenLM/Qwen3-ASR/blob/main/examples/example_qwen3_asr_transformers.py) for the transformers backend.
+
+#### vLLM Backend
+
+If you want the fastest inference speed with Qwen3-ASR, we strongly recommend using the vLLM backend by initializing the model with `Qwen3ASRModel.LLM(...)`. Example code is provided below. Note that you must install it via `pip install -U qwen-asr[vllm]`. If you want the model to output timestamps, it’s best to install FlashAttention via `pip install -U flash-attn --no-build-isolation` to speed up inference for the forced aligner model. Remember to wrap your code under `if __name__ == '__main__':` to avoid the `spawn` error described in [vLLM Troubleshooting](https://docs.vllm.ai/en/latest/usage/troubleshooting/#python-multiprocessing).
+
+```python
+import torch
+from qwen_asr import Qwen3ASRModel
+
+if __name__ == '__main__':
+    model = Qwen3ASRModel.LLM(
+        model="Qwen/Qwen3-ASR-1.7B",
+        gpu_memory_utilization=0.7,
+        max_inference_batch_size=128, # Batch size limit for inference. -1 means unlimited. Smaller values can help avoid OOM.
+        max_new_tokens=4096, # Maximum number of tokens to generate. Set a larger value for long audio input.
+        forced_aligner="Qwen/Qwen3-ForcedAligner-0.6B",
+        forced_aligner_kwargs=dict(
+            dtype=torch.bfloat16,
+            device_map="cuda:0",
+            # attn_implementation="flash_attention_2",
+        ),
+    )
+
+    results = model.transcribe(
+        audio=[
+        "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-ASR-Repo/asr_zh.wav",
+        "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-ASR-Repo/asr_en.wav",
+        ],
+        language=["Chinese", "English"], # can also be set to None for automatic language detection
+        return_time_stamps=True,
+    )
+
+    for r in results:
+        print(r.language, r.text, r.time_stamps[0])
+```
+
+For more detailed usage examples, please refer to the [example code](https://github.com/QwenLM/Qwen3-ASR/blob/main/examples/example_qwen3_asr_vllm.py) for the vLLM backend. In addition, you can start a vLLM server via the `qwen-asr-serve` command, which is a wrapper around `vllm serve`. You can pass any arguments supported by `vllm serve`, for example:
+
+```bash
+qwen-asr-serve Qwen/Qwen3-ASR-1.7B --gpu-memory-utilization 0.8 --host 0.0.0.0 --port 8000
+```
+
+And send requests to the server via:
+
+```python
+import requests
+
+url = "http://localhost:8000/v1/chat/completions"
+headers = {"Content-Type": "application/json"}
+
+data = {
+    "messages": [
+        {
+            "role": "user",
+            "content": [
+                {
+                    "type": "audio_url",
+                    "audio_url": {
+                        "url": "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-ASR-Repo/asr_en.wav"
+                    },
+                }
+            ],
+        }
+    ]
+}
+
+response = requests.post(url, headers=headers, json=data, timeout=300)
+response.raise_for_status()
+content = response.json()['choices'][0]['message']['content']
+print(content)
+
+# parse ASR output if you want
+from qwen_asr import parse_asr_output
+language, text = parse_asr_output(content)
+print(language)
+print(text)
+```
+
+#### Streaming Inference
+
+Qwen3-ASR fully supports streaming inference. Currently, streaming inference is only available with the vLLM backend. Note that streaming inference does not support batch inference or returning timestamps. Please refer to the [example code](https://github.com/QwenLM/Qwen3-ASR/blob/main/examples/example_qwen3_asr_vllm_streaming.py) for details. You can also launch a streaming web demo through the [guide](#streaming-demo) to experience Qwen3-ASR’s streaming transcription capabilities. 
+
+#### ForcedAligner Usage
+
+`Qwen3-ForcedAligner-0.6B` can align text–speech pairs and return word or character level timestamps. Here is an example of using the forced aligner directly:
+
+```python
+import torch
+from qwen_asr import Qwen3ForcedAligner
+
+model = Qwen3ForcedAligner.from_pretrained(
+    "Qwen/Qwen3-ForcedAligner-0.6B",
+    dtype=torch.bfloat16,
+    device_map="cuda:0",
+    # attn_implementation="flash_attention_2",
+)
+
+results = model.align(
+    audio="https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-ASR-Repo/asr_zh.wav",
+    text="甚至出现交易几乎停滞的情况。",
+    language="Chinese",
+)
+
+print(results[0])
+print(results[0][0].text, results[0][0].start_time, results[0][0].end_time)
+```
+
+In addition, the forced aligner supports local paths / URLs / base64 data / `(np.ndarray, sr)` inputs and batch inference. Please refer to the [example code](https://github.com/QwenLM/Qwen3-ASR/blob/main/examples/example_qwen3_forced_aligner.py) for details.
+
+### DashScope API Usage
+
+To further explore Qwen3-ASR, we encourage you to try our DashScope API for a faster and more efficient experience. For detailed API information and documentation, please refer to the following:
+
+| API Description | API Documentation (Mainland China) | API Documentation (International) |
+|------------------|-----------------------------------|------------------------------------|
+| Real-time API for Qwen3-ASR. | [https://help.aliyun.com/zh/model-studio/qwen-real-time-speech-recognition](https://help.aliyun.com/zh/model-studio/qwen-real-time-speech-recognition) | [https://www.alibabacloud.com/help/en/model-studio/qwen-real-time-speech-recognition](https://www.alibabacloud.com/help/en/model-studio/qwen-real-time-speech-recognition) |
+| FileTrans API for Qwen3-ASR. | [https://help.aliyun.com/zh/model-studio/qwen-speech-recognition](https://help.aliyun.com/zh/model-studio/qwen-speech-recognition) | [https://www.alibabacloud.com/help/en/model-studio/qwen-speech-recognition](https://www.alibabacloud.com/help/en/model-studio/qwen-speech-recognition) |
+
+
+## Launch Local Web UI Demo
+
+### Gradio Demo
+
+To launch the Qwen3-ASR web UI gradio demo, install the `qwen-asr` package and run `qwen-asr-demo`. Use the command below for help:
+
+```bash
+qwen-asr-demo --help
+```
+
+To launch the demo, you can use the following commands:
+
+```bash
+# Transformers backend
+qwen-asr-demo \
+  --asr-checkpoint Qwen/Qwen3-ASR-1.7B \
+  --backend transformers \
+  --cuda-visible-devices 0 \
+  --ip 0.0.0.0 --port 8000
+
+# Transformers backend + Forced Aligner (enable timestamps)
+qwen-asr-demo \
+  --asr-checkpoint Qwen/Qwen3-ASR-1.7B \
+  --aligner-checkpoint Qwen/Qwen3-ForcedAligner-0.6B \
+  --backend transformers \
+  --cuda-visible-devices 0 \
+  --backend-kwargs '{"device_map":"cuda:0","dtype":"bfloat16","max_inference_batch_size":8,"max_new_tokens":256}' \
+  --aligner-kwargs '{"device_map":"cuda:0","dtype":"bfloat16"}' \
+  --ip 0.0.0.0 --port 8000
+
+# vLLM backend + Forced Aligner (enable timestamps)
+qwen-asr-demo \
+  --asr-checkpoint Qwen/Qwen3-ASR-1.7B \
+  --aligner-checkpoint Qwen/Qwen3-ForcedAligner-0.6B \
+  --backend vllm \
+  --cuda-visible-devices 0 \
+  --backend-kwargs '{"gpu_memory_utilization":0.7,"max_inference_batch_size":8,"max_new_tokens":2048}' \
+  --aligner-kwargs '{"device_map":"cuda:0","dtype":"bfloat16"}' \
+  --ip 0.0.0.0 --port 8000
+```
+
+Then open `http://<your-ip>:8000`, or access it via port forwarding in tools like VS Code.
+
+#### Backend Notes
+
+This demo supports two backends: transformers and vLLM. All backend-specific initialization parameters should be passed via `--backend-kwargs` as a JSON dict. If not provided, the demo will use sensible defaults.
+
+```bash
+# Example: override transformers init args without flash attention
+--backend-kwargs '{"device_map":"cuda:0","dtype":"bfloat16"}'
+
+# Example: override vLLM init args with 65% GPU memory
+--backend-kwargs '{"gpu_memory_utilization":0.65}'
+```
+
+#### CUDA Device Notes
+
+Because vLLM does not follow `cuda:0` style device selection, this demo selects GPUs by setting `CUDA_VISIBLE_DEVICES` via `--cuda-visible-devices`.
+
+```bash
+# Use GPU 0
+--cuda-visible-devices 0
+
+# Use GPU 1
+--cuda-visible-devices 1
+```
+
+#### Timestamps Notes
+
+Timestamps are only available when `--aligner-checkpoint` is provided. If you launch the demo without a forced aligner, the timestamps UI will be hidden automatically.
+
+```bash
+# No forced aligner
+qwen-asr-demo --asr-checkpoint Qwen/Qwen3-ASR-1.7B
+
+# With forced aligner
+qwen-asr-demo \
+  --asr-checkpoint Qwen/Qwen3-ASR-1.7B \
+  --aligner-checkpoint Qwen/Qwen3-ForcedAligner-0.6B
+```
+
+#### HTTPS Notes
+
+To avoid browser microphone permission issues after deploying the server, it is recommended/required to run the gradio service over HTTPS (especially when accessed remotely or behind modern browsers/gateways). Use `--ssl-certfile` and `--ssl-keyfile` to enable HTTPS. First, generate a private key and a self-signed certificate (valid for 365 days):
+
+```bash
+openssl req -x509 -newkey rsa:2048 \
+  -keyout key.pem -out cert.pem \
+  -days 365 -nodes \
+  -subj "/CN=localhost"
+```
+
+Then run the demo with HTTPS:
+
+```bash
+qwen-asr-demo \
+  --asr-checkpoint Qwen/Qwen3-ASR-1.7B \
+  --backend transformers \
+  --cuda-visible-devices 0 \
+  --ip 0.0.0.0 --port 8000 \
+  --ssl-certfile cert.pem \
+  --ssl-keyfile key.pem \
+  --no-ssl-verify
+```
+
+Then open `https://<your-ip>:8000` to use it. If your browser shows a warning, that’s expected for self-signed certificates. For production, use a real certificate.
+
+### Streaming Demo
+
+To experience Qwen3-ASR’s streaming transcription capability in a web UI, we provide a minimal Flask-based streaming demo. The demo captures microphone audio in the browser, resamples it to 16,000 Hz, and continuously pushes PCM chunks to the model. Run the demo with the following command:
+
+```bash
+qwen-asr-demo-streaming \
+  --asr-model-path Qwen/Qwen3-ASR-1.7B \
+  --host 0.0.0.0 \
+  --port 8000 \
+  --gpu-memory-utilization 0.9
+```
+
+Then open `http://<your-ip>:8000`, or access it via port forwarding in tools like VS Code.
+
+## Deployment with vLLM
+
+vLLM officially provides day-0 model support for Qwen3-ASR for efficient inference. 
+
+### Installation
+You can run Qwen3-ASR with vLLM nightly wheel or docker image. To install the nightly version of vLLM, we recommend using `uv` as the environment manager
+```bash
+uv venv
+source .venv/bin/activate
+uv pip install -U vllm --pre \
+    --extra-index-url https://wheels.vllm.ai/nightly/cu129 \
+    --extra-index-url https://download.pytorch.org/whl/cu129 \
+    --index-strategy unsafe-best-match
+uv pip install "vllm[audio]" # For additional audio dependencies
+```
+
+### Online Serving
+You can easily deploy Qwen3-ASR with vLLM by running the following command
+```bash
+vllm serve Qwen/Qwen3-ASR-1.7B
+```
+After the model server is successfully deployed, you can interact with it in multiple ways.
+
+#### Using OpenAI SDK
+```python
+import base64
+import httpx
+from openai import OpenAI
+
+# Initialize client
+client = OpenAI(
+    base_url="http://localhost:8000/v1",
+    api_key="EMPTY"
+)
+
+# Create multimodal chat completion request
+response = client.chat.completions.create(
+    model="Qwen/Qwen3-ASR-1.7B",
+    messages=[
+        {
+            "role": "user",
+            "content": [
+                {
+                    "type": "audio_url",
+                    "audio_url": {
+                        {"url": "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-ASR-Repo/asr_en.wav"}
+                    }
+                }
+            ]
+        }
+    ],
+)
+
+print(response.choices[0].message.content)
+```
+This model is also supported on vLLM with OpenAI transcription API.
+```python
+import httpx
+from openai import OpenAI
+
+# Initialize client
+client = OpenAI(
+    base_url="http://localhost:8000/v1",
+    api_key="EMPTY"
+)
+audio_url = "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-ASR-Repo/asr_en.wav"
+audio_file = httpx.get(audio_url).content
+
+transcription = client.audio.transcriptions.create(
+    model="Qwen/Qwen3-ASR-1.7B",
+    file=audio_file,
+)
+
+print(transcription.text)
+```
+
+#### Using cURL
+```bash
+curl http://localhost:8000/v1/chat/completions \
+    -H "Content-Type: application/json" \
+    -d '{
+    "messages": [
+    {"role": "user", "content": [
+        {"type": "audio_url", "audio_url": {"url": "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen3-ASR-Repo/asr_en.wav"}}
+    ]}
+    ]
+    }'
+```
+
+### Offline Inference
+See the following example on using vLLM to run offline inference with Qwen3-ASR
+```python
+from vllm import LLM, SamplingParams
+from vllm.assets.audio import AudioAsset
+import base64
+import requests
+
+# Initialize the LLM
+llm = LLM(
+    model="Qwen/Qwen3-ASR-1.7B"
+)
+
+# Load audio
+audio_asset = AudioAsset("winning_call")
+
+# Create conversation with audio content
+conversation = [
+    {
+        "role": "user",
+        "content": [
+            {
+                "type": "audio_url",
+                "audio_url": {"url": audio_asset.url}
+            }
+        ]
+    }
+]
+
+sampling_params = SamplingParams(temperature=0.01, max_tokens=256)
+
+# Run inference using .chat()
+outputs = llm.chat(conversation, sampling_params=sampling_params)
+print(outputs[0].outputs[0].text)
+```
+
+
+## Docker
+
+To make it easier to use our `qwen-asr` Python package, we provide a pre-built Docker image: [qwenllm/qwen3-asr](https://hub.docker.com/r/qwenllm/qwen3-asr). You only need to install the GPU driver and download the model files to run the code. Please follow the [NVIDIA Container Toolkit installation guide](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html) to ensure Docker can access your GPU. If you are in Mainland China and have trouble reaching Docker Hub, you may use a registry mirror to accelerate image pulls.
+
+First, pull the image and start a container:
+
+```bash
+LOCAL_WORKDIR=/path/to/your/workspace
+HOST_PORT=8000
+CONTAINER_PORT=80
+docker run --gpus all --name qwen3-asr \
+    -v /var/run/docker.sock:/var/run/docker.sock -p $HOST_PORT:$CONTAINER_PORT \
+    --mount type=bind,source=$LOCAL_WORKDIR,target=/data/shared/Qwen3-ASR \
+    --shm-size=4gb \
+    -it qwenllm/qwen3-asr:latest
+```
+
+After running the command, you will enter the container’s bash shell. Your local workspace (**replace** `/path/to/your/workspace` **with the actual path**) will be mounted inside the container at `/data/shared/Qwen3-ASR`. Port `8000` on the host is mapped to port `80` in the container, so you can access services running in the container via `http://<host-ip>:8000`. Note that services inside the container must bind to `0.0.0.0` (not `127.0.0.1`) for port forwarding to work.
+
+If you exit the container, you can start it again and re-enter it with:
+
+```bash
+docker start qwen3-asr
+docker exec -it qwen3-asr bash
+```
+
+To remove the container completely, run:
+
+```bash
+docker rm -f qwen3-asr
+```
+
+
+## Evaluation
+
+During evaluation, we ran inference for all models with `dtype=torch.bfloat16` and set `max_new_tokens=1024` using vLLM. Greedy search was used for all decoding, and none of the tests specified a language parameter. The detailed evaluation results are shown below.
+
+<details>
+<summary>ASR Benchmarks on Public Datasets (WER ↓)</summary>
+
+<table>
+  <thead>
+    <tr>
+      <th colspan="2" style="text-align: left;"></th>
+      <th style="text-align: center;">GPT-4o<br>-Transcribe</th>
+      <th style="text-align: center;">Gemini-2.5<br>-Pro</th>
+      <th style="text-align: center;">Doubao-ASR</th>
+      <th style="text-align: center;">Whisper<br>-large-v3</th>
+      <th style="text-align: center;">Fun-ASR<br>-MLT-Nano</th>
+      <th style="text-align: center;">Qwen3-ASR<br>-0.6B</th>
+      <th style="text-align: center;">Qwen3-ASR<br>-1.7B</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td colspan="9" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">English (en)</td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">Librispeech<br>clean | other</td>
+      <td style="text-align: center;"><strong>1.39</strong> | 3.75</td>
+      <td style="text-align: center;">2.89 | 3.56</td>
+      <td style="text-align: center;">2.78 | 5.70</td>
+      <td style="text-align: center;">1.51 | 3.97</td>
+      <td style="text-align: center;">1.68 | 4.03</td>
+      <td style="text-align: center;">2.11 | 4.55</td>
+      <td style="text-align: center;">1.63 | <strong>3.38</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">GigaSpeech</td>
+      <td style="text-align: center;">25.50</td>
+      <td style="text-align: center;">9.37</td>
+      <td style="text-align: center;">9.55</td>
+      <td style="text-align: center;">9.76</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">8.88</td>
+      <td style="text-align: center;"><strong>8.45</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">CV-en</td>
+      <td style="text-align: center;">9.08</td>
+      <td style="text-align: center;">14.49</td>
+      <td style="text-align: center;">13.78</td>
+      <td style="text-align: center;">9.90</td>
+      <td style="text-align: center;">9.90</td>
+      <td style="text-align: center;">9.92</td>
+      <td style="text-align: center;"><strong>7.39</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">Fleurs-en</td>
+      <td style="text-align: center;"><strong>2.40</strong></td>
+      <td style="text-align: center;">2.94</td>
+      <td style="text-align: center;">6.31</td>
+      <td style="text-align: center;">4.08</td>
+      <td style="text-align: center;">5.49</td>
+      <td style="text-align: center;">4.39</td>
+      <td style="text-align: center;">3.35</td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">MLS-en</td>
+      <td style="text-align: center;">5.12</td>
+      <td style="text-align: center;"><strong>3.68</strong></td>
+      <td style="text-align: center;">7.09</td>
+      <td style="text-align: center;">4.87</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">6.00</td>
+      <td style="text-align: center;">4.58</td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">Tedlium</td>
+      <td style="text-align: center;">7.69</td>
+      <td style="text-align: center;">6.15</td>
+      <td style="text-align: center;">4.91</td>
+      <td style="text-align: center;">6.84</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>3.85<strong></td>
+      <td style="text-align: center;"><strong>4.50</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">VoxPopuli</td>
+      <td style="text-align: center;">10.29</td>
+      <td style="text-align: center;">11.36</td>
+      <td style="text-align: center;">12.12</td>
+      <td style="text-align: center;">12.05</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>9.96<strong></td>
+      <td style="text-align: center;"><strong>9.15</strong></td>
+    </tr>
+    <tr>
+      <td colspan="9" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">Chinese (zh)</td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">WenetSpeech<br>net | meeting</td>
+      <td style="text-align: center;">15.30 | 32.27</td>
+      <td style="text-align: center;">14.43 | 13.47</td>
+      <td style="text-align: center;">N/A</td>
+      <td style="text-align: center;">9.86 | 19.11</td>
+      <td style="text-align: center;">6.35 | -</td>
+      <td style="text-align: center;">5.97 | 6.88</td>
+      <td style="text-align: center;"><strong>4.97</strong> | <strong>5.88</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">AISHELL-2-test</td>
+      <td style="text-align: center;">4.24</td>
+      <td style="text-align: center;">11.62</td>
+      <td style="text-align: center;">2.85</td>
+      <td style="text-align: center;">5.06</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">3.15</td>
+      <td style="text-align: center;"><strong>2.71</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">SpeechIO</td>
+      <td style="text-align: center;">12.86</td>
+      <td style="text-align: center;">5.30</td>
+      <td style="text-align: center;">2.93</td>
+      <td style="text-align: center;">7.56</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">3.44</td>
+      <td style="text-align: center;"><strong>2.88</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">Fleurs-zh</td>
+      <td style="text-align: center;">2.44</td>
+      <td style="text-align: center;">2.71</td>
+      <td style="text-align: center;">2.69</td>
+      <td style="text-align: center;">4.09</td>
+      <td style="text-align: center;">3.51</td>
+      <td style="text-align: center;">2.88</td>
+      <td style="text-align: center;"><strong>2.41</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">CV-zh</td>
+      <td style="text-align: center;">6.32</td>
+      <td style="text-align: center;">7.70</td>
+      <td style="text-align: center;">5.95</td>
+      <td style="text-align: center;">12.91</td>
+      <td style="text-align: center;">6.20</td>
+      <td style="text-align: center;">6.89</td>
+      <td style="text-align: center;"><strong>5.35</strong></td>
+    </tr>
+    <tr>
+      <td colspan="9" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">Chinese Dialect</td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">KeSpeech</td>
+      <td style="text-align: center;">26.87</td>
+      <td style="text-align: center;">24.71</td>
+      <td style="text-align: center;">5.27</td>
+      <td style="text-align: center;">28.79</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">7.08</td>
+      <td style="text-align: center;"><strong>5.10</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">Fleurs-yue</td>
+      <td style="text-align: center;">4.98</td>
+      <td style="text-align: center;">9.43</td>
+      <td style="text-align: center;">4.98</td>
+      <td style="text-align: center;">9.18</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">5.79</td>
+      <td style="text-align: center;"><strong>3.98</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">CV-yue</td>
+      <td style="text-align: center;">11.36</td>
+      <td style="text-align: center;">18.76</td>
+      <td style="text-align: center;">13.20</td>
+      <td style="text-align: center;">16.23</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">9.50</td>
+      <td style="text-align: center;"><strong>7.57</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">CV-zh-tw</td>
+      <td style="text-align: center;">6.32</td>
+      <td style="text-align: center;">7.31</td>
+      <td style="text-align: center;">4.06</td>
+      <td style="text-align: center;">7.84</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">5.59</td>
+      <td style="text-align: center;"><strong>3.77</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">WenetSpeech-Yue<br>short | long</td>
+      <td style="text-align: center;">15.62 | 25.29</td>
+      <td style="text-align: center;">25.19 | 11.23</td>
+      <td style="text-align: center;">9.74 | 11.40</td>
+      <td style="text-align: center;">32.26 | 46.64</td>
+      <td style="text-align: center;">- | -</td>
+      <td style="text-align: center;">7.54 | 9.92</td>
+      <td style="text-align: center;"><strong>5.82</strong> | <strong>8.85</strong></td>
+    </tr>
+    <tr>
+      <td colspan="2" style="text-align: left;">WenetSpeech-Chuan<br>easy | hard</td>
+      <td style="text-align: center;">34.81 | 53.98</td>
+      <td style="text-align: center;">43.79 | 67.30</td>
+      <td style="text-align: center;"><strong>11.40<strong> | <strong>20.20</strong></td>
+      <td style="text-align: center;">14.35 | 26.80</td>
+      <td style="text-align: center;">- | -</td>
+      <td style="text-align: center;">13.92 | 24.45</td>
+      <td style="text-align: center;">11.99 | 21.63</td>
+    </tr>
+  </tbody>
+</table>
+
+</details>
+
+<details>
+<summary>ASR Benchmarks on Internal Datasets (WER ↓)</summary>
+
+<table>
+  <thead>
+    <tr>
+      <th style="text-align: left;"></th>
+      <th style="text-align: center;">GPT-4o<br>-Transcribe</th>
+      <th style="text-align: center;">Gemini-2.5<br>-Pro</th>
+      <th style="text-align: center;">Doubao-ASR</th>
+      <th style="text-align: center;">Whisper<br>-large-v3</th>
+      <th style="text-align: center;">Fun-ASR<br>-MLT-Nano</th>
+      <th style="text-align: center;">Qwen3-ASR<br>-0.6B</th>
+      <th style="text-align: center;">Qwen3-ASR<br>-1.7B</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td colspan="8" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">Accented English</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Dialog-Accented English</td>
+      <td style="text-align: center;">28.56</td>
+      <td style="text-align: center;">23.85</td>
+      <td style="text-align: center;">20.41</td>
+      <td style="text-align: center;">21.30</td>
+      <td style="text-align: center;">19.96</td>
+      <td style="text-align: center;"><strong>16.62<strong></td>
+      <td style="text-align: center;"><strong>16.07</strong></td>
+    </tr>
+    <tr>
+      <td colspan="8" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">Chinese Mandarin</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Elders&Kids</td>
+      <td style="text-align: center;">14.27</td>
+      <td style="text-align: center;">36.93</td>
+      <td style="text-align: center;">4.17</td>
+      <td style="text-align: center;">10.61</td>
+      <td style="text-align: center;">4.54</td>
+      <td style="text-align: center;">4.48</td>
+      <td style="text-align: center;"><strong>3.81</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">ExtremeNoise</td>
+      <td style="text-align: center;">36.11</td>
+      <td style="text-align: center;">29.06</td>
+      <td style="text-align: center;">17.04</td>
+      <td style="text-align: center;">63.17</td>
+      <td style="text-align: center;">36.55</td>
+      <td style="text-align: center;">17.88</td>
+      <td style="text-align: center;"><strong>16.17</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">TongueTwister</td>
+      <td style="text-align: center;">20.87</td>
+      <td style="text-align: center;">4.97</td>
+      <td style="text-align: center;">3.47</td>
+      <td style="text-align: center;">16.63</td>
+      <td style="text-align: center;">9.02</td>
+      <td style="text-align: center;">4.06</td>
+      <td style="text-align: center;"><strong>2.44</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Dialog-Mandarin</td>
+      <td style="text-align: center;">20.73</td>
+      <td style="text-align: center;">12.50</td>
+      <td style="text-align: center;">6.61</td>
+      <td style="text-align: center;">14.01</td>
+      <td style="text-align: center;">7.32</td>
+      <td style="text-align: center;">7.06</td>
+      <td style="text-align: center;"><strong>6.54</strong></td>
+    </tr>
+    <tr>
+      <td colspan="8" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">Chinese Dialect</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Dialog-Cantonese</td>
+      <td style="text-align: center;">16.05</td>
+      <td style="text-align: center;">14.98</td>
+      <td style="text-align: center;">7.56</td>
+      <td style="text-align: center;">31.04</td>
+      <td style="text-align: center;">5.85</td>
+      <td style="text-align: center;"><strong>4.80<strong></td>
+      <td style="text-align: center;"><strong>4.12</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Dialog-Chinese Dialects</td>
+      <td style="text-align: center;">45.37</td>
+      <td style="text-align: center;">47.70</td>
+      <td style="text-align: center;">19.85</td>
+      <td style="text-align: center;">44.55</td>
+      <td style="text-align: center;">19.41</td>
+      <td style="text-align: center;"><strong>18.24<strong></td>
+      <td style="text-align: center;"><strong>15.94</strong></td>
+    </tr>
+  </tbody>
+</table>
+<p><strong>Dialect coverage:</strong> Results for <em>Dialog-Accented English</em> are averaged over 16 accents, and results for <em>Dialog-Chinese Dialects</em> are averaged over 22 Chinese dialects.</p>
+
+</details>
+
+<details>
+<summary>Multilingual ASR Benchmarks (WER ↓)</summary>
+
+<table>
+  <thead>
+    <tr>
+      <th style="text-align: left;"></th>
+      <th style="text-align: center;">GLM-ASR<br>-Nano-2512</th>
+      <th style="text-align: center;">Whisper<br>-large-v3</th>
+      <th style="text-align: center;">Fun-ASR<br>-MLT-Nano</th>
+      <th style="text-align: center;">Qwen3-ASR<br>-0.6B</th>
+      <th style="text-align: center;">Qwen3-ASR<br>-1.7B</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td colspan="6" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">Open-sourced Benchmarks</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">MLS</td>
+      <td style="text-align: center;">13.32</td>
+      <td style="text-align: center;">8.62</td>
+      <td style="text-align: center;">28.70</td>
+      <td style="text-align: center;">13.19</td>
+      <td style="text-align: center;"><strong>8.55</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">CommonVoice</td>
+      <td style="text-align: center;">19.40</td>
+      <td style="text-align: center;">10.77</td>
+      <td style="text-align: center;">17.25</td>
+      <td style="text-align: center;">12.75</td>
+      <td style="text-align: center;"><strong>9.18</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">MLC-SLM</td>
+      <td style="text-align: center;">34.93</td>
+      <td style="text-align: center;">15.68</td>
+      <td style="text-align: center;">29.94</td>
+      <td style="text-align: center;">15.84</td>
+      <td style="text-align: center;"><strong>12.74</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Fleurs</td>
+      <td style="text-align: center;">16.08</td>
+      <td style="text-align: center;">5.27</td>
+      <td style="text-align: center;">10.03</td>
+      <td style="text-align: center;">7.57</td>
+      <td style="text-align: center;"><strong>4.90</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Fleurs<sup>†</sup></td>
+      <td style="text-align: center;">20.05</td>
+      <td style="text-align: center;">6.85</td>
+      <td style="text-align: center;">31.89</td>
+      <td style="text-align: center;">10.37</td>
+      <td style="text-align: center;"><strong>6.62</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Fleurs<sup>††</sup></td>
+      <td style="text-align: center;">24.83</td>
+      <td style="text-align: center;"><strong>8.16</strong></td>
+      <td style="text-align: center;">47.84</td>
+      <td style="text-align: center;">21.80</td>
+      <td style="text-align: center;">12.60</td>
+    </tr>
+    <tr>
+      <td colspan="6" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">Qwen-ASR Internal Benchmarks</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">News-Multilingual</td>
+      <td style="text-align: center;">49.40</td>
+      <td style="text-align: center;">14.80</td>
+      <td style="text-align: center;">65.07</td>
+      <td style="text-align: center;">17.39</td>
+      <td style="text-align: center;"><strong>12.80</strong></td>
+    </tr>
+  </tbody>
+</table>
+<p><strong>Language coverage:</strong> <em>MLS</em> includes 8 languages: {da, de, en, es, fr, it, pl, pt}.<br><em>CommonVoice</em> includes 13 languages: {en, zh, yue, zh_TW, ar, de, es, fr, it, ja, ko, pt, ru}.<br><em>MLC-SLM</em> includes 11 languages: {en, fr, de, it, pt, es, ja, ko, ru, th, vi}.<br><em>Fleurs</em> includes 12 languages: {en, zh, yue, ar, de, es, fr, it, ja, ko, pt, ru }.<br><em>Fleurs<sup>†</sup></em> includes 8 additional languages beyond Fleurs: {hi, id, ms, nl, pl, th, tr, vi}.<br><em>Fleurs<sup>††</sup></em> includes 10 additional languages beyond Fleurs<sup>†</sup>: {cs, da, el, fa, fi, fil, hu, mk, ro, sv}.<br><em>News-Multilingual</em> includes 15 languages: {ar, de, es, fr, hi, id, it, ja, ko, nl, pl, pt, ru, th, vi}.</p>
+
+</details>
+
+<details>
+<summary>Language Identification Accuracy (%) ↑</summary>
+
+<table>
+  <thead>
+    <tr>
+      <th style="text-align: left;"></th>
+      <th style="text-align: center;">Whisper-large-v3</th>
+      <th style="text-align: center;">Qwen3-ASR-0.6B</th>
+      <th style="text-align: center;">Qwen3-ASR-1.7B</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td style="text-align: left;">MLS</td>
+      <td style="text-align: center;"><strong>99.9</strong></td>
+      <td style="text-align: center;">99.3</td>
+      <td style="text-align: center;"><strong>99.9</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">CommonVoice</td>
+      <td style="text-align: center;">92.7</td>
+      <td style="text-align: center;"><strong>98.2<strong></td>
+      <td style="text-align: center;"><strong>98.7</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">MLC-SLM</td>
+      <td style="text-align: center;">89.2</td>
+      <td style="text-align: center;"><strong>92.7<strong></td>
+      <td style="text-align: center;"><strong>94.1</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Fleurs</td>
+      <td style="text-align: center;">94.6</td>
+      <td style="text-align: center;"><strong>97.1<strong></td>
+      <td style="text-align: center;"><strong>98.7</strong></td>
+    </tr>
+    <tr style="border-top: 1px solid #ddd;">
+      <td style="text-align: left;"><em>Avg.</em></td>
+      <td style="text-align: center;">94.1</td>
+      <td style="text-align: center;"><strong>96.8<strong></td>
+      <td style="text-align: center;"><strong>97.9</strong></td>
+    </tr>
+  </tbody>
+</table>
+<p><strong>Language coverage:</strong> The language sets follow Multilingual ASR Benchmarks. Here, Fleurs corresponds to Fleurs<sup>††</sup> in Multilingual ASR Benchmarks and covers 30 languages.</p>
+
+</details>
+
+<details>
+<summary>Singing Voice & Song Transcription (WER ↓)</summary>
+
+<table>
+  <thead>
+    <tr>
+      <th style="text-align: left;"></th>
+      <th style="text-align: center;">GPT-4o<br>-Transcribe</th>
+      <th style="text-align: center;">Gemini-2.5<br>-Pro</th>
+      <th style="text-align: center;">Doubao-ASR<br>-1.0</th>
+      <th style="text-align: center;">Whisper<br>-large-v3</th>
+      <th style="text-align: center;">Fun-ASR-MLT<br>-Nano</th>
+      <th style="text-align: center;">Qwen3-ASR<br>-1.7B</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td colspan="7" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">Singing</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">M4Singer</td>
+      <td style="text-align: center;">16.77</td>
+      <td style="text-align: center;">20.88</td>
+      <td style="text-align: center;">7.88</td>
+      <td style="text-align: center;">13.58</td>
+      <td style="text-align: center;">7.29</td>
+      <td style="text-align: center;"><strong>5.98</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">MIR-1k-vocal</td>
+      <td style="text-align: center;">11.87</td>
+      <td style="text-align: center;">9.85</td>
+      <td style="text-align: center;">6.56</td>
+      <td style="text-align: center;">11.71</td>
+      <td style="text-align: center;">8.17</td>
+      <td style="text-align: center;"><strong>6.25</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Opencpop</td>
+      <td style="text-align: center;">7.93</td>
+      <td style="text-align: center;">6.49</td>
+      <td style="text-align: center;">3.80</td>
+      <td style="text-align: center;">9.52</td>
+      <td style="text-align: center;"><strong>2.98</strong></td>
+      <td style="text-align: center;">3.08</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Popcs</td>
+      <td style="text-align: center;">32.84</td>
+      <td style="text-align: center;">15.13</td>
+      <td style="text-align: center;">8.97</td>
+      <td style="text-align: center;">13.77</td>
+      <td style="text-align: center;">9.42</td>
+      <td style="text-align: center;"><strong>8.52</strong></td>
+    </tr>
+    <tr>
+      <td colspan="7" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">Songs with BGM</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">EntireSongs-en</td>
+      <td style="text-align: center;">30.71</td>
+      <td style="text-align: center;"><strong>12.18</strong></td>
+      <td style="text-align: center;">33.51</td>
+      <td style="text-align: center;">N/A</td>
+      <td style="text-align: center;">N/A</td>
+      <td style="text-align: center;">14.60</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">EntireSongs-zh</td>
+      <td style="text-align: center;">34.86</td>
+      <td style="text-align: center;">18.68</td>
+      <td style="text-align: center;">23.99</td>
+      <td style="text-align: center;">N/A</td>
+      <td style="text-align: center;">N/A</td>
+      <td style="text-align: center;"><strong>13.91</strong></td>
+    </tr>
+  </tbody>
+</table>
+
+</details>
+
+<details>
+<summary>ASR Inference Mode Performance (WER ↓)</summary>
+
+<table>
+  <thead>
+    <tr>
+      <th style="text-align: left;">Model</th>
+      <th style="text-align: left;">Infer. Mode</th>
+      <th style="text-align: center;">Librispeech</th>
+      <th style="text-align: center;">Fleurs-en</th>
+      <th style="text-align: center;">Fleurs-zh</th>
+      <th style="text-align: center;">Avg.</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td rowspan="2" style="text-align: left; vertical-align: middle;">Qwen3-ASR-1.7B</td>
+      <td style="text-align: left;">Offline</td>
+      <td style="text-align: center;">1.63 | 3.38</td>
+      <td style="text-align: center;">3.35</td>
+      <td style="text-align: center;">2.41</td>
+      <td style="text-align: center;">2.69</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Streaming</td>
+      <td style="text-align: center;">1.95 | 4.51</td>
+      <td style="text-align: center;">4.02</td>
+      <td style="text-align: center;">2.84</td>
+      <td style="text-align: center;">3.33</td>
+    </tr>
+    <tr style="border-top: 1px solid #ddd;">
+      <td rowspan="2" style="text-align: left; vertical-align: middle;">Qwen3-ASR-0.6B</td>
+      <td style="text-align: left;">Offline</td>
+      <td style="text-align: center;">2.11 | 4.55</td>
+      <td style="text-align: center;">4.39</td>
+      <td style="text-align: center;">2.88</td>
+      <td style="text-align: center;">3.48</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Streaming</td>
+      <td style="text-align: center;">2.54 | 6.27</td>
+      <td style="text-align: center;">5.38</td>
+      <td style="text-align: center;">3.40</td>
+      <td style="text-align: center;">4.40</td>
+    </tr>
+  </tbody>
+</table>
+
+</details>
+
+<details>
+<summary>Forced Alignment Benchmarks (AAS ms ↓)</summary>
+
+<table>
+  <thead>
+    <tr>
+      <th style="text-align: left;"></th>
+      <th style="text-align: center;">Monotonic-Aligner</th>
+      <th style="text-align: center;">NFA</th>
+      <th style="text-align: center;">WhisperX</th>
+      <th style="text-align: center;">Qwen3-ForcedAligner-0.6B</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td colspan="5" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">MFA-Labeled Raw</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Chinese</td>
+      <td style="text-align: center;">161.1</td>
+      <td style="text-align: center;">109.8</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>33.1</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">English</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">107.5</td>
+      <td style="text-align: center;">92.1</td>
+      <td style="text-align: center;"><strong>37.5</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">French</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">100.7</td>
+      <td style="text-align: center;">145.3</td>
+      <td style="text-align: center;"><strong>41.7</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">German</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">122.7</td>
+      <td style="text-align: center;">165.1</td>
+      <td style="text-align: center;"><strong>46.5</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Italian</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">142.7</td>
+      <td style="text-align: center;">155.5</td>
+      <td style="text-align: center;"><strong>75.5</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Japanese</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>42.2</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Korean</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>37.2</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Portuguese</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>38.4</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Russian</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">200.7</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>40.2</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Spanish</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">124.7</td>
+      <td style="text-align: center;">108.0</td>
+      <td style="text-align: center;"><strong>36.8</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;"><em>Avg.</em></td>
+      <td style="text-align: center;">161.1</td>
+      <td style="text-align: center;">129.8</td>
+      <td style="text-align: center;">133.2</td>
+      <td style="text-align: center;"><strong>42.9</strong></td>
+    </tr>
+    <tr>
+      <td colspan="5" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">MFA-Labeled Concat-300s</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Chinese</td>
+      <td style="text-align: center;">1742.4</td>
+      <td style="text-align: center;">235.0</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>36.5</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">English</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">226.7</td>
+      <td style="text-align: center;">227.2</td>
+      <td style="text-align: center;"><strong>58.6</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">French</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">230.6</td>
+      <td style="text-align: center;">2052.2</td>
+      <td style="text-align: center;"><strong>53.4</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">German</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">220.3</td>
+      <td style="text-align: center;">993.4</td>
+      <td style="text-align: center;"><strong>62.4</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Italian</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">290.5</td>
+      <td style="text-align: center;">5719.4</td>
+      <td style="text-align: center;"><strong>81.6</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Japanese</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>81.3</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Korean</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>42.2</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Portuguese</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>50.0</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Russian</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">283.3</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>43.0</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Spanish</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">240.2</td>
+      <td style="text-align: center;">4549.9</td>
+      <td style="text-align: center;"><strong>39.6</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Cross-lingual</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>34.2</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;"><em>Avg.</em></td>
+      <td style="text-align: center;">1742.4</td>
+      <td style="text-align: center;">246.7</td>
+      <td style="text-align: center;">2708.4</td>
+      <td style="text-align: center;"><strong>52.9</strong></td>
+    </tr>
+    <tr>
+      <td colspan="5" style="text-align: left; font-style: italic; border-top: 1px solid #ddd; border-bottom: 1px solid #ddd;">Human-Labeled</td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Raw</td>
+      <td style="text-align: center;">49.9</td>
+      <td style="text-align: center;">88.6</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>27.8</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Raw-Noisy</td>
+      <td style="text-align: center;">53.3</td>
+      <td style="text-align: center;">89.5</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>41.8</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Concat-60s</td>
+      <td style="text-align: center;">51.1</td>
+      <td style="text-align: center;">86.7</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>25.3</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Concat-300s</td>
+      <td style="text-align: center;">410.8</td>
+      <td style="text-align: center;">140.0</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>24.8</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;">Concat-Cross-lingual</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>42.5</strong></td>
+    </tr>
+    <tr>
+      <td style="text-align: left;"><em>Avg.</em></td>
+      <td style="text-align: center;">141.3</td>
+      <td style="text-align: center;">101.2</td>
+      <td style="text-align: center;">-</td>
+      <td style="text-align: center;"><strong>32.4</strong></td>
+    </tr>
+  </tbody>
+</table>
+
+</details>
+
+
+## Citation
+
+If you find our paper and code useful in your research, please consider giving a star :star: and citation :pencil: :)
+
+```BibTeX
+@article{Qwen3-ASR,
+  title={Qwen3-ASR Technical Report},
+  author={Xian Shi, Xiong Wang, Zhifang Guo, Yongqi Wang, Pei Zhang, Xinyu Zhang, Zishan Guo, Hongkun Hao, Yu Xi, Baosong Yang, Jin Xu, Jingren Zhou, Junyang Lin},
+  journal={arXiv preprint arXiv:2601.21337},
+  year={2026}
+}
+```
+
+
+<br>

qwen-asr/chat_template.json

@@ -0,0 +1 @@
+{"chat_template": "{%- set ns = namespace(system_text=\"\") -%}\n{%- for m in messages -%}\n  {%- if m.role == 'system' -%}\n    {%- if m.content is string -%}\n      {%- set ns.system_text = ns.system_text + m.content -%}\n    {%- else -%}\n      {%- for c in m.content -%}\n        {%- if c.type == 'text' and (c.text is defined) -%}\n          {%- set ns.system_text = ns.system_text + c.text -%}\n        {%- endif -%}\n      {%- endfor -%}\n    {%- endif -%}\n  {%- endif -%}\n{%- endfor -%}\n\n{%- set ns2 = namespace(audio_tokens=\"\") -%}\n{%- for m in messages -%}\n  {%- if m.content is not string -%}\n    {%- for c in m.content -%}\n      {%- if c.type == 'audio' or ('audio' in c) or ('audio_url' in c) -%}\n        {%- set ns2.audio_tokens = ns2.audio_tokens + \"<|audio_start|><|audio_pad|><|audio_end|>\" -%}\n      {%- endif -%}\n    {%- endfor -%}\n  {%- endif -%}\n{%- endfor -%}\n\n{{- '<|im_start|>system\\n' + (ns.system_text if ns.system_text is string else '') + '<|im_end|>\\n' -}}\n{{- '<|im_start|>user\\n' + ns2.audio_tokens + '<|im_end|>\\n' -}}\n{%- if add_generation_prompt -%}\n{{- '<|im_start|>assistant\\n' -}}\n{%- endif -%}"}

qwen-asr/config.json

@@ -0,0 +1,221 @@
+{
+    "architectures": [
+      "Qwen3ASRForConditionalGeneration"
+    ],
+    "model_type": "qwen3_asr",
+    "support_languages": [
+      "Chinese",
+      "English",
+      "Cantonese",
+      "Arabic",
+      "German",
+      "French",
+      "Spanish",
+      "Portuguese",
+      "Indonesian",
+      "Italian",
+      "Korean",
+      "Russian",
+      "Thai",
+      "Vietnamese",
+      "Japanese",
+      "Turkish",
+      "Hindi",
+      "Malay",
+      "Dutch",
+      "Swedish",
+      "Danish",
+      "Finnish",
+      "Polish",
+      "Czech",
+      "Filipino",
+      "Persian",
+      "Greek",
+      "Romanian",
+      "Hungarian",
+      "Macedonian"
+    ],
+    "thinker_config": {
+      "model_type": "qwen3_asr",
+      "architectures": [
+        "Qwen3ASRForConditionalGeneration"
+      ],
+      "audio_config": {
+        "_name_or_path": "",
+        "activation_dropout": 0,
+        "activation_function": "gelu",
+        "add_cross_attention": false,
+        "architectures": null,
+        "attention_dropout": 0,
+        "bad_words_ids": null,
+        "begin_suppress_tokens": null,
+        "bos_token_id": null,
+        "chunk_size_feed_forward": 0,
+        "conv_chunksize": 500,
+        "cross_attention_hidden_size": null,
+        "d_model": 1024,
+        "decoder_start_token_id": null,
+        "diversity_penalty": 0.0,
+        "do_sample": false,
+        "downsample_hidden_size": 480,
+        "dropout": 0,
+        "dtype": null,
+        "early_stopping": false,
+        "encoder_attention_heads": 16,
+        "encoder_ffn_dim": 4096,
+        "encoder_layers": 24,
+        "encoder_no_repeat_ngram_size": 0,
+        "eos_token_id": null,
+        "exponential_decay_length_penalty": null,
+        "finetuning_task": null,
+        "forced_bos_token_id": null,
+        "forced_eos_token_id": null,
+        "id2label": {
+          "0": "LABEL_0",
+          "1": "LABEL_1"
+        },
+        "initializer_range": 0.02,
+        "is_decoder": false,
+        "is_encoder_decoder": false,
+        "label2id": {
+          "LABEL_0": 0,
+          "LABEL_1": 1
+        },
+        "length_penalty": 1.0,
+        "max_length": 20,
+        "max_source_positions": 1500,
+        "min_length": 0,
+        "model_type": "qwen3_asr_audio_encoder",
+        "n_window": 50,
+        "n_window_infer": 800,
+        "no_repeat_ngram_size": 0,
+        "num_beam_groups": 1,
+        "num_beams": 1,
+        "num_hidden_layers": 24,
+        "num_mel_bins": 128,
+        "num_return_sequences": 1,
+        "output_attentions": false,
+        "output_dim": 2048,
+        "output_hidden_states": false,
+        "output_scores": false,
+        "pad_token_id": null,
+        "prefix": null,
+        "problem_type": null,
+        "pruned_heads": {},
+        "remove_invalid_values": false,
+        "repetition_penalty": 1.0,
+        "return_dict": true,
+        "return_dict_in_generate": false,
+        "scale_embedding": false,
+        "sep_token_id": null,
+        "suppress_tokens": null,
+        "task_specific_params": null,
+        "temperature": 1.0,
+        "tf_legacy_loss": false,
+        "tie_encoder_decoder": false,
+        "tie_word_embeddings": true,
+        "tokenizer_class": null,
+        "top_k": 50,
+        "top_p": 1.0,
+        "torchscript": false,
+        "typical_p": 1.0,
+        "use_bfloat16": false
+      },
+      "audio_end_token_id": 151670,
+      "audio_start_token_id": 151669,
+      "audio_token_id": 151676,
+      "dtype": "bfloat16",
+      "initializer_range": 0.02,
+      "text_config": {
+        "_name_or_path": "",
+        "add_cross_attention": false,
+        "architectures": null,
+        "attention_bias": false,
+        "attention_dropout": 0.0,
+        "bad_words_ids": null,
+        "begin_suppress_tokens": null,
+        "bos_token_id": null,
+        "chunk_size_feed_forward": 0,
+        "cross_attention_hidden_size": null,
+        "decoder_start_token_id": null,
+        "diversity_penalty": 0.0,
+        "do_sample": false,
+        "dtype": null,
+        "early_stopping": false,
+        "encoder_no_repeat_ngram_size": 0,
+        "eos_token_id": null,
+        "exponential_decay_length_penalty": null,
+        "finetuning_task": null,
+        "forced_bos_token_id": null,
+        "forced_eos_token_id": null,
+        "head_dim": 128,
+        "hidden_act": "silu",
+        "hidden_size": 2048,
+        "id2label": {
+          "0": "LABEL_0",
+          "1": "LABEL_1"
+        },
+        "initializer_range": 0.02,
+        "intermediate_size": 6144,
+        "is_decoder": false,
+        "is_encoder_decoder": false,
+        "label2id": {
+          "LABEL_0": 0,
+          "LABEL_1": 1
+        },
+        "length_penalty": 1.0,
+        "max_length": 20,
+        "max_position_embeddings": 65536,
+        "min_length": 0,
+        "model_type": "qwen3",
+        "no_repeat_ngram_size": 0,
+        "num_attention_heads": 16,
+        "num_beam_groups": 1,
+        "num_beams": 1,
+        "num_hidden_layers": 28,
+        "num_key_value_heads": 8,
+        "num_return_sequences": 1,
+        "output_attentions": false,
+        "output_hidden_states": false,
+        "output_scores": false,
+        "pad_token_id": null,
+        "prefix": null,
+        "problem_type": null,
+        "pruned_heads": {},
+        "remove_invalid_values": false,
+        "repetition_penalty": 1.0,
+        "return_dict": true,
+        "return_dict_in_generate": false,
+        "rms_norm_eps": 1e-06,
+        "rope_scaling": {
+          "interleaved": true,
+          "mrope_interleaved": true,
+          "mrope_section": [
+            24,
+            20,
+            20
+          ],
+          "rope_type": "default",
+          "type": "default"
+        },
+        "rope_theta": 1000000,
+        "sep_token_id": null,
+        "suppress_tokens": null,
+        "task_specific_params": null,
+        "temperature": 1.0,
+        "tf_legacy_loss": false,
+        "tie_encoder_decoder": false,
+        "tie_word_embeddings": true,
+        "tokenizer_class": null,
+        "top_k": 50,
+        "top_p": 1.0,
+        "torchscript": false,
+        "typical_p": 1.0,
+        "use_bfloat16": false,
+        "use_cache": true,
+        "vocab_size": 151936
+      }
+    },
+    "transformers_version": "4.57.6"
+  }
+  

qwen-asr/generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "eos_token_id": [151643,151645],
+  "pad_token_id": 151643,
+  "do_sample": false,
+  "temperature": 0.000001
+}

qwen-asr/model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a4cd1f1a04d90b757dc7f7dd26254e69a013b19e80efe590a83c6a3bde8608d6
+size 4220320824

qwen-asr/model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e0b9d9e09e2e0238e7ef3cc8a484ab387e91b90f1900bedf88bc92d7929ccfc
+size 478200688

qwen-asr/model.safetensors.index.json

@@ -0,0 +1,715 @@
+{
+  "metadata": {
+    "format": "pt"
+  },
+  "weight_map": {
+    "thinker.audio_tower.conv2d1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.conv2d1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.conv2d2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.conv2d2.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.conv2d3.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.conv2d3.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.conv_out.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.0.fc1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.0.fc1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.0.fc2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.0.fc2.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.0.final_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.0.final_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.0.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.0.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.0.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.0.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.0.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.0.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.0.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.1.fc1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.1.fc1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.1.fc2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.1.fc2.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.1.final_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.1.final_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.1.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.1.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.1.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.1.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.1.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.1.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.1.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.10.fc1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.10.fc1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.10.fc2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.10.fc2.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.10.final_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.10.final_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.10.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.10.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.10.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.10.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.10.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.10.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.10.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.11.fc1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.11.fc1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.11.fc2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.11.fc2.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.11.final_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.11.final_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.11.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.11.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.11.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.11.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.11.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.11.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.11.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.12.fc1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.12.fc1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.12.fc2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.12.fc2.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.12.final_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.12.final_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.12.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.12.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.12.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.12.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.12.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.12.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.12.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.13.fc1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.13.fc1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.13.fc2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.13.fc2.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.13.final_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.13.final_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.13.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.13.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.13.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.13.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.13.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.13.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.13.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.14.fc1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.14.fc1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.14.fc2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.14.fc2.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.14.final_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.14.final_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.14.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.14.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.14.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.14.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.14.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.14.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.14.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.15.fc1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.15.fc1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.15.fc2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.15.fc2.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.15.final_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.15.final_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.15.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.15.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.15.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.15.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.15.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.15.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.15.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.16.fc1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.16.fc1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.16.fc2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.16.fc2.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.16.final_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.16.final_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.16.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.16.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.16.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.16.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.16.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.16.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.16.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.17.fc1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.17.fc1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.17.fc2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.17.fc2.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.17.final_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.17.final_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.17.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.17.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.17.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.17.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.17.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.17.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.17.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.18.fc1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.18.fc1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.18.fc2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.18.fc2.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.18.final_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.18.final_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.18.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.18.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.18.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.18.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.18.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.18.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.18.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.19.fc1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.19.fc1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.19.fc2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.19.fc2.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.19.final_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.19.final_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.19.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.19.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.19.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.19.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.19.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.19.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.19.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.2.fc1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.2.fc1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.2.fc2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.2.fc2.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.2.final_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.2.final_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.2.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.2.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.2.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.2.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.2.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.2.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.2.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.20.fc1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.20.fc1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.20.fc2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.20.fc2.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.20.final_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.20.final_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.20.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.20.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.20.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.20.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.20.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.20.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.20.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.21.fc1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.21.fc1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.21.fc2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.21.fc2.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.21.final_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.21.final_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.21.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.21.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.21.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.21.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.21.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.21.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.21.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.22.fc1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.22.fc1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.22.fc2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.22.fc2.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.22.final_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.22.final_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.22.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.22.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.22.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.22.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.22.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.22.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.22.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.23.fc1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.23.fc1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.23.fc2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.23.fc2.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.23.final_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.23.final_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.23.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.23.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.23.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.23.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.23.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.23.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.23.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.23.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.3.fc1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.3.fc1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.3.fc2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.3.fc2.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.3.final_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.3.final_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.3.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.3.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.3.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.3.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.3.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.3.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.3.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.4.fc1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.4.fc1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.4.fc2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.4.fc2.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.4.final_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.4.final_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.4.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.4.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.4.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.4.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.4.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.4.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.4.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.5.fc1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.5.fc1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.5.fc2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.5.fc2.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.5.final_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.5.final_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.5.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.5.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.5.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.5.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.5.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.5.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.5.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.6.fc1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.6.fc1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.6.fc2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.6.fc2.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.6.final_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.6.final_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.6.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.6.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.6.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.6.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.6.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.6.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.6.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.7.fc1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.7.fc1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.7.fc2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.7.fc2.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.7.final_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.7.final_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.7.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.7.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.7.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.7.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.7.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.7.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.7.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.8.fc1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.8.fc1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.8.fc2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.8.fc2.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.8.final_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.8.final_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.8.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.8.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.8.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.8.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.8.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.8.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.8.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.9.fc1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.9.fc1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.9.fc2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.9.fc2.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.9.final_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.9.final_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.9.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.9.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.9.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.9.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.9.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.9.self_attn_layer_norm.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.layers.9.self_attn_layer_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.ln_post.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.ln_post.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.proj1.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.proj1.weight": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.proj2.bias": "model-00001-of-00002.safetensors",
+    "thinker.audio_tower.proj2.weight": "model-00001-of-00002.safetensors",
+    "thinker.lm_head.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.embed_tokens.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.0.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.0.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.1.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.1.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.10.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.10.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.11.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.11.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.12.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.12.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.13.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.13.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.14.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.14.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.15.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.15.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.16.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.16.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.17.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.17.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.18.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.18.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.19.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.19.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.2.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.2.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.20.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.20.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.21.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.21.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.21.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.21.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.21.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.22.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.22.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.22.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.22.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.22.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.22.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.23.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.23.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.23.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.23.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.23.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.23.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.23.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.23.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.24.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.24.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.24.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.24.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.24.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.24.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.24.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.24.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.24.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.24.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.25.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.25.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.25.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.25.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.25.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.25.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.25.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.25.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.25.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.25.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.25.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.26.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.26.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.26.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.26.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.26.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.26.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.26.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.26.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.26.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.26.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.26.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.27.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.27.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.27.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.27.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.27.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.27.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.27.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.27.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.27.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.27.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.27.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.3.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.3.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.4.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.4.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "thinker.model.layers.5.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.5.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.5.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.5.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.5.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.5.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.5.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.5.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.5.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.6.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.6.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.6.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.6.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.6.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.6.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.6.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.6.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.6.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.6.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.6.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.7.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.7.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.7.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.7.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.7.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.7.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.7.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.7.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.7.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.7.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.7.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.8.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.8.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.8.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.8.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.8.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.8.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.8.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.8.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.8.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.8.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.8.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.9.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.9.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.9.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.9.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.9.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.9.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.9.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.9.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.9.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.9.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.layers.9.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "thinker.model.norm.weight": "model-00002-of-00002.safetensors"
+  }
+}

qwen-asr/preprocessor_config.json

@@ -0,0 +1,14 @@
+{
+  "chunk_length": 30,
+  "dither": 0.0,
+  "feature_extractor_type": "WhisperFeatureExtractor",
+  "feature_size": 128,
+  "hop_length": 160,
+  "n_fft": 400,
+  "n_samples": 480000,
+  "nb_max_frames": 3000,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "processor_class": "Qwen3ASRProcessor",
+  "return_attention_mask": true
+}

qwen-asr/tokenizer_config.json

@@ -0,0 +1,549 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151665": {
+      "content": "<tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151666": {
+      "content": "</tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151667": {
+      "content": "<think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151668": {
+      "content": "</think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151669": {
+      "content": "<|audio_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151670": {
+      "content": "<|audio_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151671": {
+      "content": "<tts_pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151672": {
+      "content": "<tts_text_bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151673": {
+      "content": "<tts_text_eod>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151674": {
+      "content": "<tts_text_bos_single>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151675": {
+      "content": "<non_speech>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151676": {
+      "content": "<|audio_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151677": {
+      "content": "<blank1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151678": {
+      "content": "<blank2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151679": {
+      "content": "<blank3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151680": {
+      "content": "<blank4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151681": {
+      "content": "<blank5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151682": {
+      "content": "<blank6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151683": {
+      "content": "<blank7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151684": {
+      "content": "<blank8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151685": {
+      "content": "<blank9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151686": {
+      "content": "<blank10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151687": {
+      "content": "<blank11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151688": {
+      "content": "<blank12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151689": {
+      "content": "<blank13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151690": {
+      "content": "<blank14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151691": {
+      "content": "<blank15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151692": {
+      "content": "<blank16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151693": {
+      "content": "<blank17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151694": {
+      "content": "<blank18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151695": {
+      "content": "<blank19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151696": {
+      "content": "<blank20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151697": {
+      "content": "<blank21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151698": {
+      "content": "<blank22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151699": {
+      "content": "<blank23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151700": {
+      "content": "<blank24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151701": {
+      "content": "<blank25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151702": {
+      "content": "<blank26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151703": {
+      "content": "<blank27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151704": {
+      "content": "<asr_text>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>",
+    "<|audio_start|>",
+    "<|audio_end|>",
+    "<tts_pad>",
+    "<tts_text_bos>",
+    "<tts_text_bos_single>",
+    "<|audio_pad|>"
+  ],
+  "audio_bos_token": "<|audio_start|>",
+  "audio_eos_token": "<|audio_end|>",
+  "audio_token": "<|audio_pad|>",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {
+    "audio_bos_token": "<|audio_start|>",
+    "audio_eos_token": "<|audio_end|>",
+    "audio_token": "<|audio_pad|>",
+    "image_token": "<|image_pad|>",
+    "video_token": "<|video_pad|>",
+    "vision_bos_token": "<|vision_start|>",
+    "vision_eos_token": "<|vision_end|>"
+  },
+  "image_token": "<|image_pad|>",
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "processor_class": "Qwen3ASRProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null,
+  "video_token": "<|video_pad|>",
+  "vision_bos_token": "<|vision_start|>",
+  "vision_eos_token": "<|vision_end|>"
+}

rnnoise/LICENSE

@@ -0,0 +1,31 @@
+Copyright (c) 2017, Mozilla
+Copyright (c) 2007-2017, Jean-Marc Valin
+Copyright (c) 2005-2017, Xiph.Org Foundation
+Copyright (c) 2003-2004, Mark Borgerding
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions
+are met:
+
+- Redistributions of source code must retain the above copyright
+notice, this list of conditions and the following disclaimer.
+
+- Redistributions in binary form must reproduce the above copyright
+notice, this list of conditions and the following disclaimer in the
+documentation and/or other materials provided with the distribution.
+
+- Neither the name of the Xiph.Org Foundation nor the names of its
+contributors may be used to endorse or promote products derived from
+this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+A PARTICULAR PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE FOUNDATION
+OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

rnnoise/PROVENANCE.txt

@@ -0,0 +1,2 @@
+Original model: https://raw.githubusercontent.com/richardpl/arnndn-models/0fda24c46d78f0207d820bb970fbe85c1971b39c/std.rnnn
+License: bsd-3-clause

tesseract-eng/LICENSE

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

tesseract-eng/PROVENANCE.txt

@@ -0,0 +1,2 @@
+Original model: https://cdn.jsdelivr.net/npm/@tesseract.js-data/eng/4.0.0_best_int/eng.traineddata.gz
+License: apache-2.0

tesseract-eng/eng.traineddata.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45b4cb346724ac1774f1c36f42f182b887bcdb28ebe63e6fff90ac41f3fcff91
+size 2952873

unlimited-ocr/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Baidu
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

unlimited-ocr/PROVENANCE.txt

@@ -0,0 +1,4 @@
+Original model: baidu/Unlimited-OCR
+Upstream revision: 8249d20b2822a7d8a4c22c8d8c9069b5ac36bda3
+License: mit
+Distributed for SnapTop without changing the upstream license.

unlimited-ocr/README.md

@@ -0,0 +1,335 @@
+---
+pipeline_tag: image-text-to-text
+language:
+- multilingual
+tags:
+- baidu
+- vision-language
+- ocr
+- custom_code
+license: mit
+library_name: transformers
+---
+<p align="center">
+  <img src="assets/baidu.png" width="55%" alt="Baidu Inc." />
+</p>
+
+<hr>
+
+<h1 align="center">Unlimited OCR Works</h1>
+
+<div align="center">
+
+  <a href="https://trendshift.io/repositories/62053?utm_source=trendshift-badge&amp;utm_medium=badge&amp;utm_campaign=badge-trendshift-62053" target="_blank" rel="noopener noreferrer"><img src="https://trendshift.io/api/badge/trendshift/repositories/62053/daily" alt="baidu%2FUnlimited-OCR | Trendshift" width="250" height="55"/></a>
+  
+  <a href="https://github.com/baidu/Unlimited-OCR">
+    <img alt="GitHub" src="https://img.shields.io/badge/GitHub-Code-181717?logo=github&logoColor=white" />
+  </a>
+  <a href="https://huggingface.co/baidu/Unlimited-OCR">
+    <img alt="Hugging Face" src="https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Model-ffc107?color=ffc107&logoColor=white" />
+  </a>
+</div>
+
+<div align="center">
+    <a href="https://arxiv.org/abs/2606.23050">
+    <img alt="arXiv" src="https://img.shields.io/badge/arXiv-Unlimited OCR Works-b31b1b?logo=arxiv&logoColor=white" />
+  </a>
+  <a href="https://x.com/Baidu_Inc" target="_blank">
+    <img alt="Twitter Follow" src="https://img.shields.io/badge/Twitter-Baidu Inc.-white?logo=x&logoColor=white" />
+  </a>
+</div>
+
+<h3 align="center">Welcome the Era of One-shot Long-horizon Parsing.</h3>
+
+<p align="center">
+    <img src="assets/Unlimited-OCR.png" width="1000" alt="Unlimited OCR overview" />
+</p>
+
+
+## Release
+- [2026/07/21] 🤝 Thanks to the [ms-swift community](https://github.com/modelscope/ms-swift) for their support, our model now supports training with [ms-swift](https://github.com/modelscope/ms-swift).
+- [2026/07/03] 🤝 Thanks to the Baidu Cloud team for their support. Our model is now available on [Baidu Cloud](https://cloud.baidu.com/doc/OCR/s/fmr1p39gb).
+- [2026/06/28] 🤝 Thanks to the [vLLM community](https://github.com/vllm-project/vllm) and [Tianyu Guo](https://github.com/gty111) for their support, our model now supports vLLM inference.
+- [2026/06/24] 🤝 Thanks to [AK](https://x.com/_akhaliq) for creating a demo for us. It is now available at [Hugging Face Spaces](https://huggingface.co/spaces/baidu/Unlimited-OCR).
+- [2026/06/23] 📄 Our paper is now available on [arXiv](https://arxiv.org/abs/2606.23050).
+- [2026/06/23] 🤝 Thanks to the [ModelScope community](https://github.com/modelscope) for their support. Our model is now available at [ModelScope](https://modelscope.cn/models/PaddlePaddle/Unlimited-OCR).
+- [2026/06/22] 🚀 We present [Unlimited-OCR](https://github.com/baidu/Unlimited-OCR), aiming to push [Deepseek-OCR](https://github.com/deepseek-ai/DeepSeek-OCR) one step further.
+
+## Inference
+
+### Transformers
+Inference using Huggingface transformers on NVIDIA GPUs. Requirements tested on python 3.12.3 + CUDA12.9：
+
+```
+torch==2.10.0
+torchvision==0.25.0
+transformers==4.57.1
+Pillow==12.1.1
+matplotlib==3.10.8
+einops==0.8.2
+addict==2.4.0
+easydict==1.13
+pymupdf==1.27.2.2
+psutil==7.2.2
+```
+
+```python
+import os
+import torch
+from transformers import AutoModel, AutoTokenizer
+
+model_name = 'baidu/Unlimited-OCR'
+
+tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
+model = AutoModel.from_pretrained(
+    model_name,
+    trust_remote_code=True,
+    use_safetensors=True,
+    torch_dtype=torch.bfloat16,
+)
+model = model.eval().cuda()
+
+# ── Single image supports two configs: gundam or base ──
+# gundam: base_size=1024, image_size=640, crop_mode=True
+# base: base_size=1024, image_size=1024, crop_mode=False
+model.infer(
+    tokenizer,
+    prompt='<image>document parsing.',
+    image_file='your_image.jpg',
+    output_path='your/output/dir',
+    base_size=1024, image_size=640, crop_mode=True,
+    max_length=32768,
+    no_repeat_ngram_size=35, ngram_window=128,
+    save_results=True,
+)
+
+# ── Multi page / PDF only uses base (image_size=1024) ──
+model.infer_multi(
+    tokenizer,
+    prompt='<image>Multi page parsing.',
+    image_files=['page1.png', 'page2.png', 'page3.png'],
+    output_path='your/output/dir',
+    image_size=1024,
+    max_length=32768,
+    no_repeat_ngram_size=35, ngram_window=1024,
+    save_results=True,
+)
+
+# ── PDF (convert pages to images, then multi-page parsing) ──
+import tempfile, fitz  # PyMuPDF
+
+def pdf_to_images(pdf_path, dpi=300):
+    doc = fitz.open(pdf_path)
+    tmp_dir = tempfile.mkdtemp(prefix='pdf_ocr_')
+    mat = fitz.Matrix(dpi / 72, dpi / 72)
+    paths = []
+    for i, page in enumerate(doc):
+        out = os.path.join(tmp_dir, f'page_{i+1:04d}.png')
+        page.get_pixmap(matrix=mat).save(out)
+        paths.append(out)
+    doc.close()
+    return paths
+
+model.infer_multi(
+    tokenizer,
+    prompt='<image>Multi page parsing.',
+    image_files=pdf_to_images('your_doc.pdf', dpi=300),
+    output_path='your/output/dir',
+    image_size=1024,
+    max_length=32768,
+    no_repeat_ngram_size=35, ngram_window=1024,
+    save_results=True,
+)
+```
+
+### vLLM
+
+Please refer to the official vLLM recipe for deployment details:
+
+**Recipe:** [https://recipes.vllm.ai/baidu/Unlimited-OCR](https://recipes.vllm.ai/baidu/Unlimited-OCR)
+
+##### Docker Images
+Use the following Docker images depending on your GPU platform:
+
+**Default (CUDA 13.0):**
+```bash
+docker pull vllm/vllm-openai:unlimited-ocr
+```
+**For Hopper GPUs (CUDA 12.9)**
+```bash
+docker pull vllm/vllm-openai:unlimited-ocr-cu129
+```
+
+### SGLang
+
+Set up the environment (uv-managed virtualenv). Install the local SGLang wheel first,
+then pin `kernels==0.9.0` and install PyMuPDF for PDF-to-image conversion:
+```shell
+uv venv --python 3.12
+source .venv/bin/activate
+
+uv pip install wheel/sglang-0.0.0.dev11416+g92e8bb79e-py3-none-any.whl
+uv pip install kernels==0.11.7
+uv pip install pymupdf==1.27.2.2
+```
+
+Start the SGLang server:
+```shell
+python -m sglang.launch_server \
+    --model baidu/Unlimited-OCR \
+    --served-model-name Unlimited-OCR \
+    --attention-backend fa3 \
+    --page-size 1 \
+    --mem-fraction-static 0.8 \
+    --context-length 32768 \
+    --enable-custom-logit-processor \
+    --disable-overlap-schedule \
+    --skip-server-warmup \
+    --host 0.0.0.0 \
+    --port 10000
+```
+
+Send streaming requests to the OpenAI-compatible API:
+```python
+import base64
+import json
+import os
+import tempfile
+
+import fitz
+import requests
+from sglang.srt.sampling.custom_logit_processor import DeepseekOCRNoRepeatNGramLogitProcessor
+
+server_url = "http://127.0.0.1:10000"
+
+session = requests.Session()
+session.trust_env = False
+
+
+def pdf_to_images(pdf_path, dpi=300):
+    doc = fitz.open(pdf_path)
+    tmp_dir = tempfile.mkdtemp(prefix="pdf_ocr_")
+    mat = fitz.Matrix(dpi / 72, dpi / 72)
+    image_paths = []
+    for i, page in enumerate(doc):
+        image_path = os.path.join(tmp_dir, f"page_{i + 1:04d}.png")
+        page.get_pixmap(matrix=mat).save(image_path)
+        image_paths.append(image_path)
+    doc.close()
+    return image_paths
+
+
+def encode_image(image_path):
+    ext = os.path.splitext(image_path)[1].lower()
+    mime = "image/jpeg" if ext in (".jpg", ".jpeg") else f"image/{ext.lstrip('.')}"
+    with open(image_path, "rb") as f:
+        data = base64.b64encode(f.read()).decode("utf-8")
+    return {"type": "image_url", "image_url": {"url": f"data:{mime};base64,{data}"}}
+
+
+def build_content(prompt, image_paths):
+    return [{"type": "text", "text": prompt}] + [encode_image(path) for path in image_paths]
+
+
+def generate(prompt, image_paths, image_mode, ngram_window):
+    payload = {
+        "model": "Unlimited-OCR",
+        "messages": [{"role": "user", "content": build_content(prompt, image_paths)}],
+        "temperature": 0,
+        "skip_special_tokens": False,
+        "images_config": {"image_mode": image_mode},
+        "custom_logit_processor": DeepseekOCRNoRepeatNGramLogitProcessor.to_str(),
+        "custom_params": {
+            "ngram_size": 35,
+            "window_size": ngram_window,
+        },
+        "stream": True,
+    }
+    response = session.post(
+        f"{server_url}/v1/chat/completions",
+        headers={"Content-Type": "application/json"},
+        data=json.dumps(payload),
+        timeout=1200,
+        stream=True,
+    )
+    response.raise_for_status()
+
+    chunks = []
+    for line in response.iter_lines(chunk_size=1, decode_unicode=True):
+        if not line or not line.startswith("data: "):
+            continue
+        data = line[len("data: "):]
+        if data == "[DONE]":
+            break
+        event = json.loads(data)
+        delta = event["choices"][0].get("delta", {}).get("content", "")
+        if delta:
+            print(delta, end="", flush=True)
+            chunks.append(delta)
+    print()
+    return "".join(chunks)
+
+
+# Single image supports two configs: gundam or base. Example below uses gundam.
+generate("document parsing.", ["your_image.jpg"], image_mode="gundam", ngram_window=128)
+
+# Multi image (base only)
+generate("Multi page parsing.", ["page1.png", "page2.png"], image_mode="base", ngram_window=1024)
+
+# PDF (base only)
+generate("Multi page parsing.", pdf_to_images("your_doc.pdf", dpi=300), image_mode="base", ngram_window=1024)
+```
+
+For OmniDocBench evaluation, you need to perform the following post-processing.
+```python
+DET_RE = re.compile(r'<\|det\|>([^<\s]+)(?:\s*\[[^\]]*\])?\s*<\|/det\|>(.*)', re.DOTALL)
+
+def remove_det(raw: str) -> str:
+    """
+    Strip <|det|>type [bbox]<|/det|> markers, group lines belonging to the
+    same block with \\n, and separate different blocks with \\n\\n.
+    """
+    blocks = []
+    cur = None
+    for line in raw.splitlines():
+        line = line.rstrip()
+        if not line:
+            continue
+        m = DET_RE.match(line)
+        if m:
+            category, content = m.group(1).strip(), m.group(2).strip()
+            if category == 'image':
+                continue
+            if cur is not None:
+                blocks.append(cur)
+            cur = [content] if content else []
+            continue
+        if cur is None:
+            cur = []
+        cur.append(line)
+    if cur is not None:
+        blocks.append(cur)
+    text = '\n\n'.join('\n'.join(b) for b in blocks).strip()
+    return text
+```
+
+
+## Visualization
+
+<img src="assets/long-horizon-ocr.gif" width="100%" alt="Long-horizon OCR demo" />
+
+## Acknowledgement
+
+We would like to thank [Deepseek-OCR](https://github.com/deepseek-ai/DeepSeek-OCR), [Deepseek-OCR-2](https://github.com/deepseek-ai/DeepSeek-OCR-2), [PaddleOCR](https://github.com/PaddlePaddle/PaddleOCR) for their valuable models and ideas.
+
+## Citation
+```bibtex
+@misc{yin2026unlimitedocrworks,
+      title={Unlimited OCR Works}, 
+      author={Youyang Yin and Huanhuan Liu and YY and Qunyi Xie and Chaorun Liu and Shiqi Yang and Shaohua Wang and Zhanlong Liu and Hao Zou and Jinyue Chen and Shu Wei and Jingjing Wu and Mingxin Huang and Zhen Wu and Guibin Wang and Tengyu Du and Lei Jia},
+      year={2026},
+      eprint={2606.23050},
+      archivePrefix={arXiv},
+      primaryClass={cs.CV},
+      url={https://arxiv.org/abs/2606.23050}, 
+}

unlimited-ocr/chat_template.jinja

@@ -0,0 +1,15 @@
+{%- for message in messages %}
+    {%- if message['content'] is string %}
+        {{- message['content'] }}
+    {%- else %}
+        {%- for content in message['content'] | selectattr('type', 'equalto', 'image') %}
+            {{- '<image>' }}
+        {%- endfor %}
+        {%- for content in message['content'] | selectattr('type', 'equalto', 'text') %}
+            {{- content['text'] }}
+        {%- endfor %}
+    {%- endif %}
+    {%- if message['role'] == 'assistant' %}
+        {{- eos_token }}
+    {%- endif %}
+{%- endfor %}

unlimited-ocr/config.json

@@ -0,0 +1,121 @@
+{
+  "_name_or_path": "Unlimited-OCR",
+  "candidate_resolutions": [
+    [
+      1024,
+      1024
+    ]
+  ],
+  "global_view_pos": "head",
+  "architectures": [
+    "UnlimitedOCRForCausalLM"
+  ],
+  "auto_map": {
+    "AutoConfig": "modeling_unlimitedocr.UnlimitedOCRConfig",
+    "AutoModel": "modeling_unlimitedocr.UnlimitedOCRForCausalLM"
+  },
+  "language_config": {
+    "architectures": [
+      "DeepseekOCRForCausalLM"
+    ],
+    "auto_map": {
+      "AutoConfig": "configuration_deepseekv2.DeepseekV2Config",
+      "AutoModel": "modeling_deepseek.DeepseekV2Model",
+      "AutoModelForCausalLM": "modeling_deepseek.DeepseekV2ForCausalLM"
+    },
+    "bos_token_id": 0,
+    "eos_token_id": 1,
+    "first_k_dense_replace": 1,
+    "hidden_size": 1280,
+    "intermediate_size": 6848,
+    "kv_lora_rank": null,
+    "lm_head": true,
+    "max_position_embeddings": 32768,
+    "moe_intermediate_size": 896,
+    "n_group": 1,
+    "n_routed_experts": 64,
+    "n_shared_experts": 2,
+    "num_attention_heads": 10,
+    "num_experts_per_tok": 6,
+    "num_hidden_layers": 12,
+    "num_key_value_heads": 10,
+    "q_lora_rank": null,
+    "qk_nope_head_dim": 0,
+    "qk_rope_head_dim": 0,
+    "rm_head": false,
+    "topk_group": 1,
+    "topk_method": "greedy",
+    "torch_dtype": "bfloat16",
+    "use_mla": false,
+    "v_head_dim": 128,
+    "vocab_size": 129280,
+    "sliding_window_size": 128
+  },
+  "model_type": "unlimited_ocr",
+  "projector_config": {
+    "input_dim": 2048,
+    "model_type": "mlp_projector",
+    "n_embed": 1280,
+    "projector_type": "linear"
+  },
+  "tile_tag": "2D",
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.46.3",
+  "vision_config": {
+    "image_size": 1024,
+    "mlp_ratio": 3.7362,
+    "model_name": "deeplip_b_l",
+    "model_type": "vision",
+    "width": {
+      "clip-l-14-224": {
+        "heads": 16,
+        "image_size": 224,
+        "layers": 24,
+        "patch_size": 14,
+        "width": 1024
+      },
+      "sam_vit_b": {
+        "downsample_channels": [
+          512,
+          1024
+        ],
+        "global_attn_indexes": [
+          2,
+          5,
+          8,
+          11
+        ],
+        "heads": 12,
+        "layers": 12,
+        "width": 768
+      }
+    }
+  },
+  "bos_token_id": 0,
+  "eos_token_id": 1,
+  "first_k_dense_replace": 1,
+  "hidden_size": 1280,
+  "intermediate_size": 6848,
+  "kv_lora_rank": null,
+  "lm_head": true,
+  "max_position_embeddings": 32768,
+  "moe_intermediate_size": 896,
+  "n_group": 1,
+  "n_routed_experts": 64,
+  "n_shared_experts": 2,
+  "num_attention_heads": 10,
+  "num_experts_per_tok": 6,
+  "num_hidden_layers": 12,
+  "num_key_value_heads": 10,
+  "q_lora_rank": null,
+  "qk_nope_head_dim": 0,
+  "qk_rope_head_dim": 0,
+  "rm_head": false,
+  "topk_group": 1,
+  "topk_method": "greedy",
+  "use_mla": false,
+  "v_head_dim": 128,
+  "vocab_size": 129280,
+  "sliding_window_size": 128,
+  "sliding_window": 128
+}