{
  "axquant_version": "1.2.0",
  "calibration": null,
  "created_at": "2026-08-05T08:23:55.481291Z",
  "effective_bpw": 4.949927508451917,
  "files": [
    {
      "path": "README.md",
      "sha256": "f944a3bed85ad156433c73ce4227cc8dd7e5de07a21b281f132ff8b5bf5c0dca",
      "size_bytes": 7796
    },
    {
      "path": "axquant_plan.json",
      "sha256": "6f90c1d59273dd8c482a7b14bccc90b8f7947e244ec71a572fd30ee71fe015c1",
      "size_bytes": 337379
    },
    {
      "path": "axquant_quantizer_execution.json",
      "sha256": "226b42e3e7a08581a1c8128c52ed319d5263807fafe71b47e356f324933a8b61",
      "size_bytes": 68529
    },
    {
      "path": "axquant_runtime.json",
      "sha256": "3c218a96a5a8d62d687f22d0a91985dc9164f5c8b15770573d0eaeb3e5811897",
      "size_bytes": 1628
    },
    {
      "path": "chat_template.jinja",
      "sha256": "abf41beccbd34014b52ada4fdefaa0572201d40f85ff8114f1fa5dde0949c89a",
      "size_bytes": 1590
    },
    {
      "path": "config.json",
      "sha256": "ad3c166ac4958c8542e114ea1f0aba2546253f5c3d90d678e21de09dba4c0057",
      "size_bytes": 77944
    },
    {
      "path": "generation_config.json",
      "sha256": "46a6e46680317fe06de9a5e1b2bcc4cf2965245b4994dab4600ba4be4860f1b8",
      "size_bytes": 111
    },
    {
      "path": "model-00001-of-00003.safetensors",
      "sha256": "b6580465a51a48c0d4541aaaf3017ace61f17e7bee36c49a5e3eb5e53ce0bf27",
      "size_bytes": 5308086419
    },
    {
      "path": "model-00002-of-00003.safetensors",
      "sha256": "464f89aa6a7b57c8f826c75a9c3fea284e39eb6dbc114abba21a0d3a88083642",
      "size_bytes": 5314707223
    },
    {
      "path": "model-00003-of-00003.safetensors",
      "sha256": "9060d2f447d0203a0459a36cba4ef510e3c4eebb4cae92c8228129bc123407e5",
      "size_bytes": 3962522975
    },
    {
      "path": "model.safetensors.index.json",
      "sha256": "49f2e5cc33c02b3040dbff542839e016af6fe9b0ba8aaeea67e31df919ac715a",
      "size_bytes": 79244
    },
    {
      "path": "tokenizer.json",
      "sha256": "b76085f9923309d873994d444989f7eb6ec074b06f25b58f1e8d7b7741070949",
      "size_bytes": 17078037
    },
    {
      "path": "tokenizer_config.json",
      "sha256": "56838aafa405e68956ce4e39f3f5443a48d9587302cb044d85355eac111ab201",
      "size_bytes": 345
    }
  ],
  "format": "mlx",
  "logical_parameters": 23572403200,
  "main_logical_parameters": 23572403200,
  "main_weight_file_size_bytes": 14585316617,
  "measured_main_bpw": 4.949963393465118,
  "measured_total_bpw": 4.949963393465118,
  "mtp_acceptance_retention": null,
  "mtp_distribution": {},
  "mtp_measured_speedup": null,
  "mtp_policy": {
    "candidate_bits": [
      8,
      16
    ],
    "min_bits": 8,
    "mode": "protected",
    "optimize_for_acceptance": true,
    "preserve_external_sidecar": true,
    "protect_norms": true,
    "protect_output_head": true
  },
  "mtp_present": false,
  "mtp_weight_file_size_bytes": 0,
  "plan_sha256": "7bd8665657133cce8bda462f969b954f466519dbff00600f596688ed5e84888f",
  "profile": "general",
  "protected_weight_file_size_bytes": 0,
  "quantizer": "axquant",
  "runtime": {
    "ax_engine": {
      "decode_kernel": null,
      "fused_mtp": null,
      "kernel_evidence": "unmeasured",
      "model_manifest": "model-manifest.json",
      "preferred_group_size": 32
    },
    "compatible_runtimes": [
      {
        "compatibility_level": "B",
        "manifest": "config.json",
        "mtp_support": "none",
        "name": "mlx-lm",
        "notes": [
          "Standard backbone inference is the compatibility target.",
          "AXQuant MTP metadata may be ignored by MLX-LM."
        ],
        "standard_inference": true,
        "standard_mlx_weights": true,
        "support_level": "standard-inference"
      }
    ],
    "created_at": "2026-08-05T08:23:50.819054Z",
    "kv_cache": null,
    "memory_policy": {
      "kv_cache_precision": "runtime-default",
      "mtp_buffers": "not-required",
      "prefix_cache": "runtime-managed",
      "unified_memory_safety_margin": "benchmark-required"
    },
    "mtp": {
      "acceptance_retention": null,
      "detected": false,
      "draft_tokens": null,
      "enabled_by_default": false,
      "head_precision": null,
      "measured_speedup": null,
      "optimized": false,
      "recommended_temperature_max": null,
      "sidecar_file": null,
      "verification_mode": null
    },
    "optimization_scope": "text-path",
    "primary_runtime": {
      "compatibility_level": "A",
      "manifest": "model-manifest.json",
      "mtp_support": "none",
      "name": "ax-engine",
      "notes": [
        "Runtime claims require a passing AX Engine doctor and benchmark report."
      ],
      "standard_inference": true,
      "standard_mlx_weights": true,
      "support_level": "optimized"
    },
    "schema_version": "axquant.runtime.v1"
  },
  "schema_version": "axquant.artifact.v2",
  "software_versions": {
    "ax_engine": null,
    "axquant": "1.2.0",
    "mlx": "0.32.0",
    "mlx_lm": "0.31.3",
    "pydantic": "2.13.4",
    "python": "3.13.14",
    "safetensors": "0.8.0"
  },
  "source_model": {
    "architecture": "MistralForCausalLM",
    "format": "mlx",
    "local_path": null,
    "model_id": "mistralai/Devstral-Small-2505",
    "revision": "c2a9d81a2989af566682b4cecc828c84556076c5"
  },
  "target_class": "4bit",
  "weight_distribution": {
    "4bit": {
      "fraction": 0.9430439065288005,
      "parameters": 22229811200
    },
    "8bit": {
      "fraction": 0.028469250008416623,
      "parameters": 671088640
    },
    "bf16": {
      "fraction": 0.028486843462782786,
      "parameters": 671503360
    }
  },
  "weight_file_size_bytes": 14585316617
}