finbase0530 Youssofal commited on
Commit
0aa3682
·
0 Parent(s):

Duplicate from Youssofal/Qwen3.6-27B-MTPLX-Optimized-Speed

Browse files

Co-authored-by: Youssof Altoukhi <Youssofal@users.noreply.huggingface.co>

.gitattributes ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ *.7z filter=lfs diff=lfs merge=lfs -text
2
+ *.arrow filter=lfs diff=lfs merge=lfs -text
3
+ *.bin filter=lfs diff=lfs merge=lfs -text
4
+ *.bz2 filter=lfs diff=lfs merge=lfs -text
5
+ *.ckpt filter=lfs diff=lfs merge=lfs -text
6
+ *.ftz filter=lfs diff=lfs merge=lfs -text
7
+ *.gz filter=lfs diff=lfs merge=lfs -text
8
+ *.h5 filter=lfs diff=lfs merge=lfs -text
9
+ *.joblib filter=lfs diff=lfs merge=lfs -text
10
+ *.lfs.* filter=lfs diff=lfs merge=lfs -text
11
+ *.mlmodel filter=lfs diff=lfs merge=lfs -text
12
+ *.model filter=lfs diff=lfs merge=lfs -text
13
+ *.msgpack filter=lfs diff=lfs merge=lfs -text
14
+ *.npy filter=lfs diff=lfs merge=lfs -text
15
+ *.npz filter=lfs diff=lfs merge=lfs -text
16
+ *.onnx filter=lfs diff=lfs merge=lfs -text
17
+ *.ot filter=lfs diff=lfs merge=lfs -text
18
+ *.parquet filter=lfs diff=lfs merge=lfs -text
19
+ *.pb filter=lfs diff=lfs merge=lfs -text
20
+ *.pickle filter=lfs diff=lfs merge=lfs -text
21
+ *.pkl filter=lfs diff=lfs merge=lfs -text
22
+ *.pt filter=lfs diff=lfs merge=lfs -text
23
+ *.pth filter=lfs diff=lfs merge=lfs -text
24
+ *.rar filter=lfs diff=lfs merge=lfs -text
25
+ *.safetensors filter=lfs diff=lfs merge=lfs -text
26
+ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
27
+ *.tar.* filter=lfs diff=lfs merge=lfs -text
28
+ *.tar filter=lfs diff=lfs merge=lfs -text
29
+ *.tflite filter=lfs diff=lfs merge=lfs -text
30
+ *.tgz filter=lfs diff=lfs merge=lfs -text
31
+ *.wasm filter=lfs diff=lfs merge=lfs -text
32
+ *.xz filter=lfs diff=lfs merge=lfs -text
33
+ *.zip filter=lfs diff=lfs merge=lfs -text
34
+ *.zst filter=lfs diff=lfs merge=lfs -text
35
+ *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
LICENSE ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ Apache License
3
+ Version 2.0, January 2004
4
+ http://www.apache.org/licenses/
5
+
6
+ TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
7
+
8
+ 1. Definitions.
9
+
10
+ "License" shall mean the terms and conditions for use, reproduction,
11
+ and distribution as defined by Sections 1 through 9 of this document.
12
+
13
+ "Licensor" shall mean the copyright owner or entity authorized by
14
+ the copyright owner that is granting the License.
15
+
16
+ "Legal Entity" shall mean the union of the acting entity and all
17
+ other entities that control, are controlled by, or are under common
18
+ control with that entity. For the purposes of this definition,
19
+ "control" means (i) the power, direct or indirect, to cause the
20
+ direction or management of such entity, whether by contract or
21
+ otherwise, or (ii) ownership of fifty percent (50%) or more of the
22
+ outstanding shares, or (iii) beneficial ownership of such entity.
23
+
24
+ "You" (or "Your") shall mean an individual or Legal Entity
25
+ exercising permissions granted by this License.
26
+
27
+ "Source" form shall mean the preferred form for making modifications,
28
+ including but not limited to software source code, documentation
29
+ source, and configuration files.
30
+
31
+ "Object" form shall mean any form resulting from mechanical
32
+ transformation or translation of a Source form, including but
33
+ not limited to compiled object code, generated documentation,
34
+ and conversions to other media types.
35
+
36
+ "Work" shall mean the work of authorship, whether in Source or
37
+ Object form, made available under the License, as indicated by a
38
+ copyright notice that is included in or attached to the work
39
+ (an example is provided in the Appendix below).
40
+
41
+ "Derivative Works" shall mean any work, whether in Source or Object
42
+ form, that is based on (or derived from) the Work and for which the
43
+ editorial revisions, annotations, elaborations, or other modifications
44
+ represent, as a whole, an original work of authorship. For the purposes
45
+ of this License, Derivative Works shall not include works that remain
46
+ separable from, or merely link (or bind by name) to the interfaces of,
47
+ the Work and Derivative Works thereof.
48
+
49
+ "Contribution" shall mean any work of authorship, including
50
+ the original version of the Work and any modifications or additions
51
+ to that Work or Derivative Works thereof, that is intentionally
52
+ submitted to Licensor for inclusion in the Work by the copyright owner
53
+ or by an individual or Legal Entity authorized to submit on behalf of
54
+ the copyright owner. For the purposes of this definition, "submitted"
55
+ means any form of electronic, verbal, or written communication sent
56
+ to the Licensor or its representatives, including but not limited to
57
+ communication on electronic mailing lists, source code control systems,
58
+ and issue tracking systems that are managed by, or on behalf of, the
59
+ Licensor for the purpose of discussing and improving the Work, but
60
+ excluding communication that is conspicuously marked or otherwise
61
+ designated in writing by the copyright owner as "Not a Contribution."
62
+
63
+ "Contributor" shall mean Licensor and any individual or Legal Entity
64
+ on behalf of whom a Contribution has been received by Licensor and
65
+ subsequently incorporated within the Work.
66
+
67
+ 2. Grant of Copyright License. Subject to the terms and conditions of
68
+ this License, each Contributor hereby grants to You a perpetual,
69
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
70
+ copyright license to reproduce, prepare Derivative Works of,
71
+ publicly display, publicly perform, sublicense, and distribute the
72
+ Work and such Derivative Works in Source or Object form.
73
+
74
+ 3. Grant of Patent License. Subject to the terms and conditions of
75
+ this License, each Contributor hereby grants to You a perpetual,
76
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
77
+ (except as stated in this section) patent license to make, have made,
78
+ use, offer to sell, sell, import, and otherwise transfer the Work,
79
+ where such license applies only to those patent claims licensable
80
+ by such Contributor that are necessarily infringed by their
81
+ Contribution(s) alone or by combination of their Contribution(s)
82
+ with the Work to which such Contribution(s) was submitted. If You
83
+ institute patent litigation against any entity (including a
84
+ cross-claim or counterclaim in a lawsuit) alleging that the Work
85
+ or a Contribution incorporated within the Work constitutes direct
86
+ or contributory patent infringement, then any patent licenses
87
+ granted to You under this License for that Work shall terminate
88
+ as of the date such litigation is filed.
89
+
90
+ 4. Redistribution. You may reproduce and distribute copies of the
91
+ Work or Derivative Works thereof in any medium, with or without
92
+ modifications, and in Source or Object form, provided that You
93
+ meet the following conditions:
94
+
95
+ (a) You must give any other recipients of the Work or
96
+ Derivative Works a copy of this License; and
97
+
98
+ (b) You must cause any modified files to carry prominent notices
99
+ stating that You changed the files; and
100
+
101
+ (c) You must retain, in the Source form of any Derivative Works
102
+ that You distribute, all copyright, patent, trademark, and
103
+ attribution notices from the Source form of the Work,
104
+ excluding those notices that do not pertain to any part of
105
+ the Derivative Works; and
106
+
107
+ (d) If the Work includes a "NOTICE" text file as part of its
108
+ distribution, then any Derivative Works that You distribute must
109
+ include a readable copy of the attribution notices contained
110
+ within such NOTICE file, excluding those notices that do not
111
+ pertain to any part of the Derivative Works, in at least one
112
+ of the following places: within a NOTICE text file distributed
113
+ as part of the Derivative Works; within the Source form or
114
+ documentation, if provided along with the Derivative Works; or,
115
+ within a display generated by the Derivative Works, if and
116
+ wherever such third-party notices normally appear. The contents
117
+ of the NOTICE file are for informational purposes only and
118
+ do not modify the License. You may add Your own attribution
119
+ notices within Derivative Works that You distribute, alongside
120
+ or as an addendum to the NOTICE text from the Work, provided
121
+ that such additional attribution notices cannot be construed
122
+ as modifying the License.
123
+
124
+ You may add Your own copyright statement to Your modifications and
125
+ may provide additional or different license terms and conditions
126
+ for use, reproduction, or distribution of Your modifications, or
127
+ for any such Derivative Works as a whole, provided Your use,
128
+ reproduction, and distribution of the Work otherwise complies with
129
+ the conditions stated in this License.
130
+
131
+ 5. Submission of Contributions. Unless You explicitly state otherwise,
132
+ any Contribution intentionally submitted for inclusion in the Work
133
+ by You to the Licensor shall be under the terms and conditions of
134
+ this License, without any additional terms or conditions.
135
+ Notwithstanding the above, nothing herein shall supersede or modify
136
+ the terms of any separate license agreement you may have executed
137
+ with Licensor regarding such Contributions.
138
+
139
+ 6. Trademarks. This License does not grant permission to use the trade
140
+ names, trademarks, service marks, or product names of the Licensor,
141
+ except as required for reasonable and customary use in describing the
142
+ origin of the Work and reproducing the content of the NOTICE file.
143
+
144
+ 7. Disclaimer of Warranty. Unless required by applicable law or
145
+ agreed to in writing, Licensor provides the Work (and each
146
+ Contributor provides its Contributions) on an "AS IS" BASIS,
147
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
148
+ implied, including, without limitation, any warranties or conditions
149
+ of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
150
+ PARTICULAR PURPOSE. You are solely responsible for determining the
151
+ appropriateness of using or redistributing the Work and assume any
152
+ risks associated with Your exercise of permissions under this License.
153
+
154
+ 8. Limitation of Liability. In no event and under no legal theory,
155
+ whether in tort (including negligence), contract, or otherwise,
156
+ unless required by applicable law (such as deliberate and grossly
157
+ negligent acts) or agreed to in writing, shall any Contributor be
158
+ liable to You for damages, including any direct, indirect, special,
159
+ incidental, or consequential damages of any character arising as a
160
+ result of this License or out of the use or inability to use the
161
+ Work (including but not limited to damages for loss of goodwill,
162
+ work stoppage, computer failure or malfunction, or any and all
163
+ other commercial damages or losses), even if such Contributor
164
+ has been advised of the possibility of such damages.
165
+
166
+ 9. Accepting Warranty or Additional Liability. While redistributing
167
+ the Work or Derivative Works thereof, You may choose to offer,
168
+ and charge a fee for, acceptance of support, warranty, indemnity,
169
+ or other liability obligations and/or rights consistent with this
170
+ License. However, in accepting such obligations, You may act only
171
+ on Your own behalf and on Your sole responsibility, not on behalf
172
+ of any other Contributor, and only if You agree to indemnify,
173
+ defend, and hold each Contributor harmless for any liability
174
+ incurred by, or claims asserted against, such Contributor by reason
175
+ of your accepting any such warranty or additional liability.
176
+
177
+ END OF TERMS AND CONDITIONS
178
+
179
+ APPENDIX: How to apply the Apache License to your work.
180
+
181
+ To apply the Apache License to your work, attach the following
182
+ boilerplate notice, with the fields enclosed by brackets "[]"
183
+ replaced with your own identifying information. (Don't include
184
+ the brackets!) The text should be enclosed in the appropriate
185
+ comment syntax for the file format. We also recommend that a
186
+ file or class name and description of purpose be included on the
187
+ same "printed page" as the copyright notice for easier
188
+ identification within third-party archives.
189
+
190
+ Copyright 2026 Alibaba Cloud
191
+
192
+ Licensed under the Apache License, Version 2.0 (the "License");
193
+ you may not use this file except in compliance with the License.
194
+ You may obtain a copy of the License at
195
+
196
+ http://www.apache.org/licenses/LICENSE-2.0
197
+
198
+ Unless required by applicable law or agreed to in writing, software
199
+ distributed under the License is distributed on an "AS IS" BASIS,
200
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
201
+ See the License for the specific language governing permissions and
202
+ limitations under the License.
MTPLX_PUBLISH_MANIFEST.json ADDED
@@ -0,0 +1,175 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "created_at": "2026-05-03T16:59:16.607393+00:00",
3
+ "files": [
4
+ {
5
+ "materialization": "clonefile",
6
+ "name": "chat_template.jinja",
7
+ "resolved_source_path": "/Users/youssof/.lmstudio/models/mlx-community/Qwen3.6-27B-4bit/chat_template.jinja",
8
+ "same_inode_as_source": false,
9
+ "size_bytes": 7764,
10
+ "source_path": "/Users/youssof/Documents/MTPLX/models/Qwen3.6-27B-MTPLX-Flat4-CyanKiwiMTP/chat_template.jinja"
11
+ },
12
+ {
13
+ "materialization": "clonefile",
14
+ "name": "config.json",
15
+ "resolved_source_path": "/Users/youssof/Documents/MTPLX/models/Qwen3.6-27B-MTPLX-Flat4-CyanKiwiMTP/config.json",
16
+ "same_inode_as_source": false,
17
+ "size_bytes": 5541,
18
+ "source_path": "/Users/youssof/Documents/MTPLX/models/Qwen3.6-27B-MTPLX-Flat4-CyanKiwiMTP/config.json"
19
+ },
20
+ {
21
+ "materialization": "clonefile",
22
+ "name": "configuration.json",
23
+ "resolved_source_path": "/Users/youssof/.lmstudio/models/mlx-community/Qwen3.6-27B-4bit/configuration.json",
24
+ "same_inode_as_source": false,
25
+ "size_bytes": 51,
26
+ "source_path": "/Users/youssof/Documents/MTPLX/models/Qwen3.6-27B-MTPLX-Flat4-CyanKiwiMTP/configuration.json"
27
+ },
28
+ {
29
+ "materialization": "clonefile",
30
+ "name": "generation_config.json",
31
+ "resolved_source_path": "/Users/youssof/.lmstudio/models/mlx-community/Qwen3.6-27B-4bit/generation_config.json",
32
+ "same_inode_as_source": false,
33
+ "size_bytes": 202,
34
+ "source_path": "/Users/youssof/Documents/MTPLX/models/Qwen3.6-27B-MTPLX-Flat4-CyanKiwiMTP/generation_config.json"
35
+ },
36
+ {
37
+ "materialization": "clonefile",
38
+ "name": "model-00001-of-00003.safetensors",
39
+ "resolved_source_path": "/Users/youssof/.lmstudio/models/mlx-community/Qwen3.6-27B-4bit/model-00001-of-00003.safetensors",
40
+ "same_inode_as_source": false,
41
+ "size_bytes": 5343268752,
42
+ "source_path": "/Users/youssof/Documents/MTPLX/models/Qwen3.6-27B-MTPLX-Flat4-CyanKiwiMTP/model-00001-of-00003.safetensors"
43
+ },
44
+ {
45
+ "materialization": "clonefile",
46
+ "name": "model-00002-of-00003.safetensors",
47
+ "resolved_source_path": "/Users/youssof/.lmstudio/models/mlx-community/Qwen3.6-27B-4bit/model-00002-of-00003.safetensors",
48
+ "same_inode_as_source": false,
49
+ "size_bytes": 5354185100,
50
+ "source_path": "/Users/youssof/Documents/MTPLX/models/Qwen3.6-27B-MTPLX-Flat4-CyanKiwiMTP/model-00002-of-00003.safetensors"
51
+ },
52
+ {
53
+ "materialization": "clonefile",
54
+ "name": "model-00003-of-00003.safetensors",
55
+ "resolved_source_path": "/Users/youssof/.lmstudio/models/mlx-community/Qwen3.6-27B-4bit/model-00003-of-00003.safetensors",
56
+ "same_inode_as_source": false,
57
+ "size_bytes": 5357087747,
58
+ "source_path": "/Users/youssof/Documents/MTPLX/models/Qwen3.6-27B-MTPLX-Flat4-CyanKiwiMTP/model-00003-of-00003.safetensors"
59
+ },
60
+ {
61
+ "materialization": "clonefile",
62
+ "name": "model.safetensors.index.json",
63
+ "resolved_source_path": "/Users/youssof/.lmstudio/models/mlx-community/Qwen3.6-27B-4bit/model.safetensors.index.json",
64
+ "same_inode_as_source": false,
65
+ "size_bytes": 218281,
66
+ "source_path": "/Users/youssof/Documents/MTPLX/models/Qwen3.6-27B-MTPLX-Flat4-CyanKiwiMTP/model.safetensors.index.json"
67
+ },
68
+ {
69
+ "materialization": "clonefile",
70
+ "name": "mtp.safetensors",
71
+ "resolved_source_path": "/Users/youssof/Documents/MTPLX/models/Qwen3.6-27B-MTPLX-CyanKiwi-Packed-BF16-INT4-v3/mtp.safetensors",
72
+ "same_inode_as_source": false,
73
+ "size_bytes": 337565787,
74
+ "source_path": "/Users/youssof/Documents/MTPLX/models/Qwen3.6-27B-MTPLX-Flat4-CyanKiwiMTP/mtp.safetensors"
75
+ },
76
+ {
77
+ "materialization": "clonefile",
78
+ "name": "mtplx_runtime.json",
79
+ "resolved_source_path": "/Users/youssof/Documents/MTPLX/models/Qwen3.6-27B-MTPLX-Flat4-CyanKiwiMTP/mtplx_runtime.json",
80
+ "same_inode_as_source": false,
81
+ "size_bytes": 2007,
82
+ "source_path": "/Users/youssof/Documents/MTPLX/models/Qwen3.6-27B-MTPLX-Flat4-CyanKiwiMTP/mtplx_runtime.json"
83
+ },
84
+ {
85
+ "materialization": "clonefile",
86
+ "name": "preprocessor_config.json",
87
+ "resolved_source_path": "/Users/youssof/.lmstudio/models/mlx-community/Qwen3.6-27B-4bit/preprocessor_config.json",
88
+ "same_inode_as_source": false,
89
+ "size_bytes": 390,
90
+ "source_path": "/Users/youssof/Documents/MTPLX/models/Qwen3.6-27B-MTPLX-Flat4-CyanKiwiMTP/preprocessor_config.json"
91
+ },
92
+ {
93
+ "materialization": "clonefile",
94
+ "name": "processor_config.json",
95
+ "resolved_source_path": "/Users/youssof/.lmstudio/models/mlx-community/Qwen3.6-27B-4bit/processor_config.json",
96
+ "same_inode_as_source": false,
97
+ "size_bytes": 1312,
98
+ "source_path": "/Users/youssof/Documents/MTPLX/models/Qwen3.6-27B-MTPLX-Flat4-CyanKiwiMTP/processor_config.json"
99
+ },
100
+ {
101
+ "materialization": "clonefile",
102
+ "name": "tokenizer.json",
103
+ "resolved_source_path": "/Users/youssof/.lmstudio/models/mlx-community/Qwen3.6-27B-4bit/tokenizer.json",
104
+ "same_inode_as_source": false,
105
+ "size_bytes": 19989343,
106
+ "source_path": "/Users/youssof/Documents/MTPLX/models/Qwen3.6-27B-MTPLX-Flat4-CyanKiwiMTP/tokenizer.json"
107
+ },
108
+ {
109
+ "materialization": "clonefile",
110
+ "name": "tokenizer_config.json",
111
+ "resolved_source_path": "/Users/youssof/.lmstudio/models/mlx-community/Qwen3.6-27B-4bit/tokenizer_config.json",
112
+ "same_inode_as_source": false,
113
+ "size_bytes": 1139,
114
+ "source_path": "/Users/youssof/Documents/MTPLX/models/Qwen3.6-27B-MTPLX-Flat4-CyanKiwiMTP/tokenizer_config.json"
115
+ },
116
+ {
117
+ "materialization": "clonefile",
118
+ "name": "video_preprocessor_config.json",
119
+ "resolved_source_path": "/Users/youssof/.lmstudio/models/mlx-community/Qwen3.6-27B-4bit/video_preprocessor_config.json",
120
+ "same_inode_as_source": false,
121
+ "size_bytes": 385,
122
+ "source_path": "/Users/youssof/Documents/MTPLX/models/Qwen3.6-27B-MTPLX-Flat4-CyanKiwiMTP/video_preprocessor_config.json"
123
+ },
124
+ {
125
+ "materialization": "clonefile",
126
+ "name": "vocab.json",
127
+ "resolved_source_path": "/Users/youssof/.lmstudio/models/mlx-community/Qwen3.6-27B-4bit/vocab.json",
128
+ "same_inode_as_source": false,
129
+ "size_bytes": 6722759,
130
+ "source_path": "/Users/youssof/Documents/MTPLX/models/Qwen3.6-27B-MTPLX-Flat4-CyanKiwiMTP/vocab.json"
131
+ },
132
+ {
133
+ "materialization": "clonefile",
134
+ "name": "LICENSE",
135
+ "resolved_source_path": "/Users/youssof/Documents/MTPLX/models/Qwen3.6-27B-MTPLX-CyanKiwi-Packed-BF16-INT4-v2/LICENSE",
136
+ "same_inode_as_source": false,
137
+ "size_bytes": 11343,
138
+ "source_path": "/Users/youssof/Documents/MTPLX/models/Qwen3.6-27B-MTPLX-CyanKiwi-Packed-BF16-INT4-v3/LICENSE"
139
+ },
140
+ {
141
+ "materialization": "generated",
142
+ "name": "README.md",
143
+ "resolved_source_path": null,
144
+ "same_inode_as_source": false,
145
+ "size_bytes": 3041,
146
+ "source_path": null
147
+ },
148
+ {
149
+ "materialization": "generated",
150
+ "name": "NOTICE",
151
+ "resolved_source_path": null,
152
+ "same_inode_as_source": false,
153
+ "size_bytes": 460,
154
+ "source_path": null
155
+ }
156
+ ],
157
+ "include_hashes": false,
158
+ "repo_id": "Youssofal/Qwen3.6-27B-MTPLX-Optimized-Speed",
159
+ "size_bytes": 16419071404,
160
+ "source_provenance": {
161
+ "base_model": "Qwen/Qwen3.6-27B",
162
+ "base_model_revision": "6a9e13bd6fc8f0983b9b99948120bc37f49c13e9",
163
+ "license": "Apache-2.0",
164
+ "mtp_sidecar_revision": "8cc526ddcfdd53346f94fb9b5458ca8495a63218",
165
+ "mtp_sidecar_source": "cyankiwi/Qwen3.6-27B-AWQ-BF16-INT4"
166
+ },
167
+ "source_wrapper": "/Users/youssof/Documents/MTPLX/models/Qwen3.6-27B-MTPLX-Flat4-CyanKiwiMTP",
168
+ "staging_dir": "/Users/youssof/Documents/MTPLX/hf-staging/Qwen3.6-27B-MTPLX-Optimized-Speed",
169
+ "upload_policy": {
170
+ "recommended_command_after_approval": "HF_HOME=/Users/youssof/.cache/huggingface-mtplx hf upload-large-folder Youssofal/Qwen3.6-27B-MTPLX-Optimized-Speed /Users/youssof/Documents/MTPLX/hf-staging/Qwen3.6-27B-MTPLX-Optimized-Speed --private",
171
+ "requires_explicit_green_light": true,
172
+ "uploaded": false
173
+ },
174
+ "weight_size_bytes": 16392107386
175
+ }
NOTICE ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ MTPLX model artifact notice
2
+
3
+ This staged artifact contains modified/quantized weights derived from:
4
+
5
+ - Qwen/Qwen3.6-27B at revision 6a9e13bd6fc8f0983b9b99948120bc37f49c13e9, licensed Apache-2.0.
6
+ - cyankiwi/Qwen3.6-27B-AWQ-BF16-INT4 at revision 8cc526ddcfdd53346f94fb9b5458ca8495a63218, licensed Apache-2.0.
7
+
8
+ The MTPLX runtime contract, model card, and staging manifest are authored for
9
+ MTPLX v0.1.0-preview. The Apache-2.0 license text is included as LICENSE.
README.md ADDED
@@ -0,0 +1,115 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: mlx
3
+ license: apache-2.0
4
+ base_model:
5
+ - Qwen/Qwen3.6-27B
6
+ pipeline_tag: text-generation
7
+ tags:
8
+ - mlx
9
+ - apple-silicon
10
+ - speculative-decoding
11
+ - qwen
12
+ - qwen3
13
+ - qwen3-next
14
+ - mtp
15
+ - mtplx
16
+ - local-ai
17
+ ---
18
+
19
+ # Qwen3.6-27B MTPLX Optimized Speed
20
+
21
+ ## Run this with MTPLX
22
+
23
+ **MTPLX** is an MLX-native runtime for native Multi-Token-Prediction speculative decoding on Apple Silicon. Up to **2.24× faster decode** at real coding temperatures (`temp=0.6 / top_p=0.95 / top_k=20`) using the model's own built-in MTP heads — no external drafter, no greedy hack.
24
+
25
+ ```bash
26
+ pip install mtplx
27
+ mtplx start
28
+ ```
29
+
30
+ **Project:** [github.com/youssofal/MTPLX](https://github.com/youssofal/MTPLX)
31
+
32
+ **Other MTPLX checkpoints:**
33
+
34
+ - [Qwen3.6-27B-MTPLX-Optimized](https://huggingface.co/Youssofal/Qwen3.6-27B-MTPLX-Optimized) — verified default (GDN8-Speed4 trunk + CyanKiwi INT4 MTP)
35
+ - [Qwen3.5-4B-MTPLX-Optimized-Speed](https://huggingface.co/Youssofal/Qwen3.5-4B-MTPLX-Optimized-Speed) — small 4-bit speed-test
36
+ - [Qwen3.5-4B-Optimized-MTPLX](https://huggingface.co/Youssofal/Qwen3.5-4B-Optimized-MTPLX) — small 8-bit
37
+
38
+ ---
39
+
40
+ This is the speed checkpoint for MTPLX v0.1.0-preview. It combines the
41
+ MLXCommunity flat 4-bit MLX trunk with the calibrated CyanKiwi INT4 MTP sidecar
42
+ and embeds `mtplx_runtime.json` so MTPLX can run it as the optimized speed
43
+ variant.
44
+
45
+ Target repository: `Youssofal/Qwen3.6-27B-MTPLX-Optimized-Speed`
46
+
47
+ ## Runtime
48
+
49
+ ```bash
50
+ mtplx pull Youssofal/Qwen3.6-27B-MTPLX-Optimized-Speed
51
+ mtplx run "hello" --model Youssofal/Qwen3.6-27B-MTPLX-Optimized-Speed
52
+ mtplx serve --model Youssofal/Qwen3.6-27B-MTPLX-Optimized-Speed --port 8000 --no-stats-footer
53
+ ```
54
+
55
+ MTPLX v0.1.0-preview supports Qwen3-Next-MTP only. This model is verified for
56
+ the `qwen3-next-mtp` backend.
57
+
58
+ ## Sources And Attribution
59
+
60
+ | Component | Source | Revision | License |
61
+ |---|---|---:|---|
62
+ | Base model | `Qwen/Qwen3.6-27B` | `6a9e13bd6fc8f0983b9b99948120bc37f49c13e9` | Apache-2.0 |
63
+ | Calibrated MTP sidecar source | `cyankiwi/Qwen3.6-27B-AWQ-BF16-INT4` | `8cc526ddcfdd53346f94fb9b5458ca8495a63218` | Apache-2.0 |
64
+ | MTPLX conversion and runtime contract | `youssofal/mtplx` | v0.1.0-preview.1 | Apache-2.0 |
65
+
66
+ The trunk is the MLXCommunity flat 4-bit affine conversion of
67
+ `Qwen/Qwen3.6-27B`. The MTP sidecar is derived from the downloaded CyanKiwi
68
+ compressed-tensors checkpoint at the revision listed above.
69
+
70
+ ## Verification
71
+
72
+ `mtplx_runtime.json` records:
73
+
74
+ - architecture: `qwen3-next-mtp`
75
+ - maximum MTP depth: `3`
76
+ - recommended profile: `performance-cold`
77
+ - recommended draft-only LM head: `3-bit affine, group_size=64`
78
+ - recommended draft sampler: temperature `0.70`, top-p `0.95`, top-k `20`
79
+ - exactness gate: `Phase 0H paged-verifier smoke`
80
+ - exactness max absolute diff: `0.0`
81
+ - verified hardware: `Apple M5 Max, 128 GB unified memory`
82
+ - verified timestamp: `2026-05-03T23:07:00+0100`
83
+
84
+ Target sampler used for the recorded contract: temperature `0.6`,
85
+ top-p `0.95`, top-k `20`, with thinking mode disabled. The draft proposal
86
+ sampler is intentionally slightly hotter at temperature `0.70`; target
87
+ verification and residual correction remain exact.
88
+
89
+ ## Performance Honesty
90
+
91
+ This is the default speed lane. On the local Apple M5 Max fanmax
92
+ performance-cold benchmark, this artifact reached **63.056 tok/s** and
93
+ **62.886 tok/s** at depth 3 on the long-code 192-token prompt when using its
94
+ contract-recommended 3-bit draft-only LM head plus draft sampler temperature
95
+ `0.70`, with thinking mode disabled and foreground app load reduced. The
96
+ proposal change reduced the measured correction tokens from 6 to 3 and verify
97
+ calls from 50 to 49 on the recorded prompt; target sampling remains temperature
98
+ `0.6` / top-p `0.95` / top-k `20`. A full greedy diagnostic on the same cleaned
99
+ fanmax window measured **60.108 tok/s**. Earlier contract evidence measured
100
+ 61.527 and 60.900 tok/s, the earlier 3-bit draft-head contract measured 60.038
101
+ and 60.061 tok/s, and the older 4-bit draft-only LM head measured 57.668 tok/s
102
+ on the same lane.
103
+
104
+ ## Files
105
+
106
+ - `model-*.safetensors`: MLXCommunity flat 4-bit trunk shards.
107
+ - `mtp.safetensors`: calibrated prequantized INT4 MTP sidecar.
108
+ - `mtplx_runtime.json`: MTPLX verified runtime contract.
109
+ - `MTPLX_PUBLISH_MANIFEST.json`: local staging manifest with source paths,
110
+ materialization method, sizes, and optional hashes.
111
+
112
+ ## Links
113
+
114
+ - **MTPLX**: [github.com/youssofal/MTPLX](https://github.com/youssofal/MTPLX) · `pip install mtplx`
115
+ - **Base model**: [Qwen/Qwen3.6-27B](https://huggingface.co/Qwen/Qwen3.6-27B)
chat_template.jinja ADDED
@@ -0,0 +1,154 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- set image_count = namespace(value=0) %}
2
+ {%- set video_count = namespace(value=0) %}
3
+ {%- macro render_content(content, do_vision_count, is_system_content=false) %}
4
+ {%- if content is string %}
5
+ {{- content }}
6
+ {%- elif content is iterable and content is not mapping %}
7
+ {%- for item in content %}
8
+ {%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
9
+ {%- if is_system_content %}
10
+ {{- raise_exception('System message cannot contain images.') }}
11
+ {%- endif %}
12
+ {%- if do_vision_count %}
13
+ {%- set image_count.value = image_count.value + 1 %}
14
+ {%- endif %}
15
+ {%- if add_vision_id %}
16
+ {{- 'Picture ' ~ image_count.value ~ ': ' }}
17
+ {%- endif %}
18
+ {{- '<|vision_start|><|image_pad|><|vision_end|>' }}
19
+ {%- elif 'video' in item or item.type == 'video' %}
20
+ {%- if is_system_content %}
21
+ {{- raise_exception('System message cannot contain videos.') }}
22
+ {%- endif %}
23
+ {%- if do_vision_count %}
24
+ {%- set video_count.value = video_count.value + 1 %}
25
+ {%- endif %}
26
+ {%- if add_vision_id %}
27
+ {{- 'Video ' ~ video_count.value ~ ': ' }}
28
+ {%- endif %}
29
+ {{- '<|vision_start|><|video_pad|><|vision_end|>' }}
30
+ {%- elif 'text' in item %}
31
+ {{- item.text }}
32
+ {%- else %}
33
+ {{- raise_exception('Unexpected item type in content.') }}
34
+ {%- endif %}
35
+ {%- endfor %}
36
+ {%- elif content is none or content is undefined %}
37
+ {{- '' }}
38
+ {%- else %}
39
+ {{- raise_exception('Unexpected content type.') }}
40
+ {%- endif %}
41
+ {%- endmacro %}
42
+ {%- if not messages %}
43
+ {{- raise_exception('No messages provided.') }}
44
+ {%- endif %}
45
+ {%- if tools and tools is iterable and tools is not mapping %}
46
+ {{- '<|im_start|>system\n' }}
47
+ {{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
48
+ {%- for tool in tools %}
49
+ {{- "\n" }}
50
+ {{- tool | tojson }}
51
+ {%- endfor %}
52
+ {{- "\n</tools>" }}
53
+ {{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
54
+ {%- if messages[0].role == 'system' %}
55
+ {%- set content = render_content(messages[0].content, false, true)|trim %}
56
+ {%- if content %}
57
+ {{- '\n\n' + content }}
58
+ {%- endif %}
59
+ {%- endif %}
60
+ {{- '<|im_end|>\n' }}
61
+ {%- else %}
62
+ {%- if messages[0].role == 'system' %}
63
+ {%- set content = render_content(messages[0].content, false, true)|trim %}
64
+ {{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
65
+ {%- endif %}
66
+ {%- endif %}
67
+ {%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
68
+ {%- for message in messages[::-1] %}
69
+ {%- set index = (messages|length - 1) - loop.index0 %}
70
+ {%- if ns.multi_step_tool and message.role == "user" %}
71
+ {%- set content = render_content(message.content, false)|trim %}
72
+ {%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
73
+ {%- set ns.multi_step_tool = false %}
74
+ {%- set ns.last_query_index = index %}
75
+ {%- endif %}
76
+ {%- endif %}
77
+ {%- endfor %}
78
+ {%- if ns.multi_step_tool %}
79
+ {{- raise_exception('No user query found in messages.') }}
80
+ {%- endif %}
81
+ {%- for message in messages %}
82
+ {%- set content = render_content(message.content, true)|trim %}
83
+ {%- if message.role == "system" %}
84
+ {%- if not loop.first %}
85
+ {{- raise_exception('System message must be at the beginning.') }}
86
+ {%- endif %}
87
+ {%- elif message.role == "user" %}
88
+ {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
89
+ {%- elif message.role == "assistant" %}
90
+ {%- set reasoning_content = '' %}
91
+ {%- if message.reasoning_content is string %}
92
+ {%- set reasoning_content = message.reasoning_content %}
93
+ {%- else %}
94
+ {%- if '</think>' in content %}
95
+ {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
96
+ {%- set content = content.split('</think>')[-1].lstrip('\n') %}
97
+ {%- endif %}
98
+ {%- endif %}
99
+ {%- set reasoning_content = reasoning_content|trim %}
100
+ {%- if (preserve_thinking is defined and preserve_thinking is true) or (loop.index0 > ns.last_query_index) %}
101
+ {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
102
+ {%- else %}
103
+ {{- '<|im_start|>' + message.role + '\n' + content }}
104
+ {%- endif %}
105
+ {%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
106
+ {%- for tool_call in message.tool_calls %}
107
+ {%- if tool_call.function is defined %}
108
+ {%- set tool_call = tool_call.function %}
109
+ {%- endif %}
110
+ {%- if loop.first %}
111
+ {%- if content|trim %}
112
+ {{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
113
+ {%- else %}
114
+ {{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
115
+ {%- endif %}
116
+ {%- else %}
117
+ {{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
118
+ {%- endif %}
119
+ {%- if tool_call.arguments is defined %}
120
+ {%- for args_name, args_value in tool_call.arguments|items %}
121
+ {{- '<parameter=' + args_name + '>\n' }}
122
+ {%- set args_value = args_value | string if args_value is string else args_value | tojson | safe %}
123
+ {{- args_value }}
124
+ {{- '\n</parameter>\n' }}
125
+ {%- endfor %}
126
+ {%- endif %}
127
+ {{- '</function>\n</tool_call>' }}
128
+ {%- endfor %}
129
+ {%- endif %}
130
+ {{- '<|im_end|>\n' }}
131
+ {%- elif message.role == "tool" %}
132
+ {%- if loop.previtem and loop.previtem.role != "tool" %}
133
+ {{- '<|im_start|>user' }}
134
+ {%- endif %}
135
+ {{- '\n<tool_response>\n' }}
136
+ {{- content }}
137
+ {{- '\n</tool_response>' }}
138
+ {%- if not loop.last and loop.nextitem.role != "tool" %}
139
+ {{- '<|im_end|>\n' }}
140
+ {%- elif loop.last %}
141
+ {{- '<|im_end|>\n' }}
142
+ {%- endif %}
143
+ {%- else %}
144
+ {{- raise_exception('Unexpected message role.') }}
145
+ {%- endif %}
146
+ {%- endfor %}
147
+ {%- if add_generation_prompt %}
148
+ {{- '<|im_start|>assistant\n' }}
149
+ {%- if enable_thinking is defined and enable_thinking is false %}
150
+ {{- '<think>\n\n</think>\n\n' }}
151
+ {%- else %}
152
+ {{- '<think>\n' }}
153
+ {%- endif %}
154
+ {%- endif %}
config.json ADDED
@@ -0,0 +1,175 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3_5ForConditionalGeneration"
4
+ ],
5
+ "eos_token_id": [
6
+ 248046,
7
+ 248044
8
+ ],
9
+ "image_token_id": 248056,
10
+ "language_model_only": false,
11
+ "mlx_lm_extra_tensors": {
12
+ "mtp_file": "mtp.safetensors",
13
+ "mtp_tensor_count": 15
14
+ },
15
+ "model_type": "qwen3_5",
16
+ "mtplx_mtp_quantization": {
17
+ "bits": 4,
18
+ "description": "Load calibrated CyanKiwi MTP layer linears as packed MLX INT4; keep mtp.fc and MTP norms BF16.",
19
+ "group_size": 32,
20
+ "mode": "affine",
21
+ "policy": "cyankiwi",
22
+ "prequantized": true
23
+ },
24
+ "mtplx_policy": {
25
+ "base_trunk": "models/Qwen3.6-27B-MLXCommunity-4bit-mtp-graft",
26
+ "description": "Legacy flat 4-bit MLXCommunity trunk with calibrated CyanKiwi prequantized INT4 MTP sidecar; speed-ceiling probe only, not production default.",
27
+ "main_policy": "mlx-community-flat-4bit",
28
+ "mtp_policy": "cyankiwi-calibrated-int4-prequantized",
29
+ "mtp_sidecar_source": "models/Qwen3.6-27B-MTPLX-CyanKiwi-Packed-BF16-INT4-v3",
30
+ "name": "mlx-community-4bit-cyankiwi-mtp-speed-probe",
31
+ "source": "Qwen/Qwen3.6-27B"
32
+ },
33
+ "quantization": {
34
+ "bits": 4,
35
+ "group_size": 64,
36
+ "mode": "affine"
37
+ },
38
+ "quantization_config": {
39
+ "bits": 4,
40
+ "group_size": 64,
41
+ "mode": "affine"
42
+ },
43
+ "text_config": {
44
+ "attention_bias": false,
45
+ "attention_dropout": 0.0,
46
+ "attn_output_gate": true,
47
+ "bos_token_id": 248044,
48
+ "dtype": "bfloat16",
49
+ "eos_token_id": 248044,
50
+ "full_attention_interval": 4,
51
+ "head_dim": 256,
52
+ "hidden_act": "silu",
53
+ "hidden_size": 5120,
54
+ "initializer_range": 0.02,
55
+ "intermediate_size": 17408,
56
+ "layer_types": [
57
+ "linear_attention",
58
+ "linear_attention",
59
+ "linear_attention",
60
+ "full_attention",
61
+ "linear_attention",
62
+ "linear_attention",
63
+ "linear_attention",
64
+ "full_attention",
65
+ "linear_attention",
66
+ "linear_attention",
67
+ "linear_attention",
68
+ "full_attention",
69
+ "linear_attention",
70
+ "linear_attention",
71
+ "linear_attention",
72
+ "full_attention",
73
+ "linear_attention",
74
+ "linear_attention",
75
+ "linear_attention",
76
+ "full_attention",
77
+ "linear_attention",
78
+ "linear_attention",
79
+ "linear_attention",
80
+ "full_attention",
81
+ "linear_attention",
82
+ "linear_attention",
83
+ "linear_attention",
84
+ "full_attention",
85
+ "linear_attention",
86
+ "linear_attention",
87
+ "linear_attention",
88
+ "full_attention",
89
+ "linear_attention",
90
+ "linear_attention",
91
+ "linear_attention",
92
+ "full_attention",
93
+ "linear_attention",
94
+ "linear_attention",
95
+ "linear_attention",
96
+ "full_attention",
97
+ "linear_attention",
98
+ "linear_attention",
99
+ "linear_attention",
100
+ "full_attention",
101
+ "linear_attention",
102
+ "linear_attention",
103
+ "linear_attention",
104
+ "full_attention",
105
+ "linear_attention",
106
+ "linear_attention",
107
+ "linear_attention",
108
+ "full_attention",
109
+ "linear_attention",
110
+ "linear_attention",
111
+ "linear_attention",
112
+ "full_attention",
113
+ "linear_attention",
114
+ "linear_attention",
115
+ "linear_attention",
116
+ "full_attention",
117
+ "linear_attention",
118
+ "linear_attention",
119
+ "linear_attention",
120
+ "full_attention"
121
+ ],
122
+ "linear_conv_kernel_dim": 4,
123
+ "linear_key_head_dim": 128,
124
+ "linear_num_key_heads": 16,
125
+ "linear_num_value_heads": 48,
126
+ "linear_value_head_dim": 128,
127
+ "mamba_ssm_dtype": "float32",
128
+ "max_position_embeddings": 262144,
129
+ "model_type": "qwen3_5_text",
130
+ "mtp_num_hidden_layers": 1,
131
+ "mtp_use_dedicated_embeddings": false,
132
+ "num_attention_heads": 24,
133
+ "num_hidden_layers": 64,
134
+ "num_key_value_heads": 4,
135
+ "output_gate_type": "swish",
136
+ "pad_token_id": null,
137
+ "partial_rotary_factor": 0.25,
138
+ "rms_norm_eps": 1e-06,
139
+ "rope_parameters": {
140
+ "mrope_interleaved": true,
141
+ "mrope_section": [
142
+ 11,
143
+ 11,
144
+ 10
145
+ ],
146
+ "partial_rotary_factor": 0.25,
147
+ "rope_theta": 10000000,
148
+ "rope_type": "default"
149
+ },
150
+ "tie_word_embeddings": false,
151
+ "use_cache": true,
152
+ "vocab_size": 248320
153
+ },
154
+ "tie_word_embeddings": false,
155
+ "transformers_version": "4.57.1",
156
+ "video_token_id": 248057,
157
+ "vision_config": {
158
+ "deepstack_visual_indexes": [],
159
+ "depth": 27,
160
+ "hidden_act": "gelu_pytorch_tanh",
161
+ "hidden_size": 1152,
162
+ "in_channels": 3,
163
+ "initializer_range": 0.02,
164
+ "intermediate_size": 4304,
165
+ "model_type": "qwen3_5",
166
+ "num_heads": 16,
167
+ "num_position_embeddings": 2304,
168
+ "out_hidden_size": 5120,
169
+ "patch_size": 16,
170
+ "spatial_merge_size": 2,
171
+ "temporal_patch_size": 2
172
+ },
173
+ "vision_end_token_id": 248054,
174
+ "vision_start_token_id": 248053
175
+ }
configuration.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"framework":"Pytorch","task":"image-text-to-text"}
generation_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 248044,
3
+ "do_sample": true,
4
+ "eos_token_id": [
5
+ 248046,
6
+ 248044
7
+ ],
8
+ "pad_token_id": 248044,
9
+ "temperature": 1.0,
10
+ "top_k": 20,
11
+ "top_p": 0.95
12
+ }
model-00001-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2689680915661f040c50c35244d08b336def279e509e1ca11873f8dd1b0e7ce0
3
+ size 5343268752
model-00002-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a46183727b2c5cd16613fd8395f5e9e7cc4ad679644558cc87c9406fe334463d
3
+ size 5354185100
model-00003-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac23bf70b1f239a040921d6f93770d74176fd435dbf44e42317053d06c68d702
3
+ size 5357087747
model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
mtp.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ebc44ee9ddcab9d812b5b1f31399c35c721878a95b7bbf57a318ca28bbb83efb
3
+ size 337565787
mtplx_runtime.json ADDED
@@ -0,0 +1,131 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "arch_id": "qwen3-next-mtp",
3
+ "artifact_role": "maximum-speed-flat4-candidate",
4
+ "base_trunk": "mlx-community/Qwen3.6-27B-4bit",
5
+ "exactness_baseline": {
6
+ "attention_impl": "mlx_vector_paged",
7
+ "context": 64,
8
+ "gate": "phase0h-paged-verifier-exactness",
9
+ "max_abs_diff": 0.0,
10
+ "mode": "decode-from-stock-prefix",
11
+ "sample_agreement": 1.0,
12
+ "status": "passed",
13
+ "topk_overlap_ratio": 1.0,
14
+ "total_variation": 0.0,
15
+ "verify_tokens": 4
16
+ },
17
+ "mtp_depth_max": 3,
18
+ "mtp_sidecar": "Qwen3.6-27B-MTPLX-CyanKiwi-Packed-BF16-INT4-v3",
19
+ "mtplx_version": "0.1.0-preview",
20
+ "recommended_draft_lm_head": {
21
+ "bits": 3,
22
+ "group_size": 64,
23
+ "mode": "affine"
24
+ },
25
+ "recommended_draft_sampler": {
26
+ "temperature": 0.7,
27
+ "top_k": 20,
28
+ "top_p": 0.95
29
+ },
30
+ "recommended_profile": "performance-cold",
31
+ "sampler": {
32
+ "temperature": 0.6,
33
+ "top_k": 20,
34
+ "top_p": 0.95
35
+ },
36
+ "speed_evidence": {
37
+ "acceptance_by_depth": [
38
+ 1.0,
39
+ 0.9795918367346939,
40
+ 0.9387755102040817
41
+ ],
42
+ "accepted_by_depth": [
43
+ 49,
44
+ 48,
45
+ 46
46
+ ],
47
+ "artifacts": [
48
+ "/Users/youssof/.mtplx/fanmax_flat4_draft_temp070_cleanapps_20260503-230229.json",
49
+ "/Users/youssof/.mtplx/fanmax_flat4_draft_temp070_cleanapps_repeat_20260503-230458.json"
50
+ ],
51
+ "correction_tokens": 3,
52
+ "depth": 3,
53
+ "draft_lm_head": "3-bit affine group64 draft-only head",
54
+ "draft_sampler": {
55
+ "temperature": 0.7,
56
+ "top_k": 20,
57
+ "top_p": 0.95
58
+ },
59
+ "drafted_by_depth": [
60
+ 49,
61
+ 49,
62
+ 49
63
+ ],
64
+ "enable_thinking": false,
65
+ "greedy_diagnostic": {
66
+ "accepted_by_depth": [
67
+ 49,
68
+ 49,
69
+ 43
70
+ ],
71
+ "artifact": "/Users/youssof/.mtplx/fanmax_flat4_full_greedy_cleanapps_20260503-230229.json",
72
+ "correction_tokens": 0,
73
+ "draft_sampler": {
74
+ "temperature": 0.0,
75
+ "top_k": 0,
76
+ "top_p": 1.0
77
+ },
78
+ "drafted_by_depth": [
79
+ 51,
80
+ 51,
81
+ 51
82
+ ],
83
+ "target_sampler": {
84
+ "temperature": 0.0,
85
+ "top_k": 0,
86
+ "top_p": 1.0
87
+ },
88
+ "tok_s": 60.108346853627076,
89
+ "verify_calls": 51
90
+ },
91
+ "older_3bit_artifacts": [
92
+ "/Users/youssof/.mtplx/fanmax_flat4_cyankiwi_draft_head_3bit_isolated_20260503.json",
93
+ "/Users/youssof/.mtplx/fanmax_flat4_cyankiwi_draft_head_3bit_repeat_20260503.json"
94
+ ],
95
+ "older_3bit_tok_s": [
96
+ 60.03822277200243,
97
+ 60.06063543439426
98
+ ],
99
+ "previous_best_artifacts": [
100
+ "/Users/youssof/.mtplx/fanmax_flat4_draft_temp070_optimized_env_20260503.json",
101
+ "/Users/youssof/.mtplx/fanmax_flat4_draft_temp070_repeat_optimized_env_20260503.json"
102
+ ],
103
+ "previous_best_tok_s": [
104
+ 61.52744855840259,
105
+ 60.89966247416089
106
+ ],
107
+ "production_cli_contract_checks": [
108
+ {
109
+ "artifact": "/Users/youssof/.mtplx/fanmax_mtplx_flat4_cyankiwi_cli_contract_perfcold_disablethinking_20260503.json",
110
+ "tok_s": 59.51602930081416
111
+ },
112
+ {
113
+ "artifact": "/Users/youssof/.mtplx/fanmax_mtplx_flat4_cyankiwi_cli_contract_perfcold_disablethinking_repeat_20260503.json",
114
+ "tok_s": 59.97165246907193
115
+ }
116
+ ],
117
+ "profile": "performance-cold",
118
+ "timestamp": "2026-05-03T23:07:00+0100",
119
+ "tok_s": [
120
+ 63.05651084048216,
121
+ 62.885564723941826
122
+ ],
123
+ "verify_calls": 49
124
+ },
125
+ "verified_on": {
126
+ "hardware": "Apple M5 Max, 128 GB unified memory",
127
+ "machine_arch": "arm64",
128
+ "model": "Qwen3.6-27B-MTPLX-Flat4-CyanKiwiMTP",
129
+ "timestamp": "2026-05-03T17:49:00+0100"
130
+ }
131
+ }
preprocessor_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "size": {
3
+ "longest_edge": 16777216,
4
+ "shortest_edge": 65536
5
+ },
6
+ "patch_size": 16,
7
+ "temporal_patch_size": 2,
8
+ "merge_size": 2,
9
+ "image_mean": [
10
+ 0.5,
11
+ 0.5,
12
+ 0.5
13
+ ],
14
+ "image_std": [
15
+ 0.5,
16
+ 0.5,
17
+ 0.5
18
+ ],
19
+ "processor_class": "Qwen3VLProcessor",
20
+ "image_processor_type": "Qwen2VLImageProcessorFast"
21
+ }
processor_config.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "image_processor": {
3
+ "do_convert_rgb": true,
4
+ "do_normalize": true,
5
+ "do_rescale": true,
6
+ "do_resize": true,
7
+ "image_mean": [
8
+ 0.5,
9
+ 0.5,
10
+ 0.5
11
+ ],
12
+ "image_processor_type": "Qwen2VLImageProcessor",
13
+ "image_std": [
14
+ 0.5,
15
+ 0.5,
16
+ 0.5
17
+ ],
18
+ "max_pixels": 16777216,
19
+ "merge_size": 2,
20
+ "min_pixels": 65536,
21
+ "patch_size": 16,
22
+ "resample": 3,
23
+ "rescale_factor": 0.00392156862745098,
24
+ "size": {
25
+ "longest_edge": 16777216,
26
+ "shortest_edge": 65536
27
+ },
28
+ "temporal_patch_size": 2
29
+ },
30
+ "processor_class": "Qwen3VLProcessor",
31
+ "video_processor": {
32
+ "data_format": "channels_first",
33
+ "default_to_square": true,
34
+ "do_convert_rgb": true,
35
+ "do_normalize": true,
36
+ "do_rescale": true,
37
+ "do_resize": true,
38
+ "do_sample_frames": true,
39
+ "fps": 2,
40
+ "image_mean": [
41
+ 0.5,
42
+ 0.5,
43
+ 0.5
44
+ ],
45
+ "image_std": [
46
+ 0.5,
47
+ 0.5,
48
+ 0.5
49
+ ],
50
+ "max_frames": 768,
51
+ "merge_size": 2,
52
+ "min_frames": 4,
53
+ "patch_size": 16,
54
+ "resample": 3,
55
+ "rescale_factor": 0.00392156862745098,
56
+ "return_metadata": false,
57
+ "size": {
58
+ "longest_edge": 25165824,
59
+ "shortest_edge": 4096
60
+ },
61
+ "temporal_patch_size": 2,
62
+ "video_processor_type": "Qwen3VLVideoProcessor"
63
+ }
64
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:87a7830d63fcf43bf241c3c5242e96e62dd3fdc29224ca26fed8ea333db72de4
3
+ size 19989343
tokenizer_config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "audio_bos_token": "<|audio_start|>",
4
+ "audio_eos_token": "<|audio_end|>",
5
+ "audio_token": "<|audio_pad|>",
6
+ "backend": "tokenizers",
7
+ "bos_token": null,
8
+ "clean_up_tokenization_spaces": false,
9
+ "eos_token": "<|im_end|>",
10
+ "errors": "replace",
11
+ "image_token": "<|image_pad|>",
12
+ "is_local": true,
13
+ "model_max_length": 262144,
14
+ "model_specific_special_tokens": {
15
+ "audio_bos_token": "<|audio_start|>",
16
+ "audio_eos_token": "<|audio_end|>",
17
+ "audio_token": "<|audio_pad|>",
18
+ "image_token": "<|image_pad|>",
19
+ "video_token": "<|video_pad|>",
20
+ "vision_bos_token": "<|vision_start|>",
21
+ "vision_eos_token": "<|vision_end|>"
22
+ },
23
+ "pad_token": "<|endoftext|>",
24
+ "pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
25
+ "processor_class": "Qwen3VLProcessor",
26
+ "split_special_tokens": false,
27
+ "tokenizer_class": "TokenizersBackend",
28
+ "unk_token": null,
29
+ "video_token": "<|video_pad|>",
30
+ "vision_bos_token": "<|vision_start|>",
31
+ "vision_eos_token": "<|vision_end|>"
32
+ }
video_preprocessor_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "size": {
3
+ "longest_edge": 25165824,
4
+ "shortest_edge": 4096
5
+ },
6
+ "patch_size": 16,
7
+ "temporal_patch_size": 2,
8
+ "merge_size": 2,
9
+ "image_mean": [
10
+ 0.5,
11
+ 0.5,
12
+ 0.5
13
+ ],
14
+ "image_std": [
15
+ 0.5,
16
+ 0.5,
17
+ 0.5
18
+ ],
19
+ "processor_class": "Qwen3VLProcessor",
20
+ "video_processor_type": "Qwen3VLVideoProcessor"
21
+ }
vocab.json ADDED
The diff for this file is too large to render. See raw diff