RubyWitch commited on
Commit
883374f
·
verified ·
1 Parent(s): 7fd9af9

Publish Holo4 NVFP4 with Qwen BF16 MTP head and provenance

Browse files
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* 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
 
 
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 +1,408 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Attribution-NonCommercial 4.0 International
2
+
3
+ =======================================================================
4
+
5
+ Creative Commons Corporation ("Creative Commons") is not a law firm and
6
+ does not provide legal services or legal advice. Distribution of
7
+ Creative Commons public licenses does not create a lawyer-client or
8
+ other relationship. Creative Commons makes its licenses and related
9
+ information available on an "as-is" basis. Creative Commons gives no
10
+ warranties regarding its licenses, any material licensed under their
11
+ terms and conditions, or any related information. Creative Commons
12
+ disclaims all liability for damages resulting from their use to the
13
+ fullest extent possible.
14
+
15
+ Using Creative Commons Public Licenses
16
+
17
+ Creative Commons public licenses provide a standard set of terms and
18
+ conditions that creators and other rights holders may use to share
19
+ original works of authorship and other material subject to copyright
20
+ and certain other rights specified in the public license below. The
21
+ following considerations are for informational purposes only, are not
22
+ exhaustive, and do not form part of our licenses.
23
+
24
+ Considerations for licensors: Our public licenses are
25
+ intended for use by those authorized to give the public
26
+ permission to use material in ways otherwise restricted by
27
+ copyright and certain other rights. Our licenses are
28
+ irrevocable. Licensors should read and understand the terms
29
+ and conditions of the license they choose before applying it.
30
+ Licensors should also secure all rights necessary before
31
+ applying our licenses so that the public can reuse the
32
+ material as expected. Licensors should clearly mark any
33
+ material not subject to the license. This includes other CC-
34
+ licensed material, or material used under an exception or
35
+ limitation to copyright. More considerations for licensors:
36
+ wiki.creativecommons.org/Considerations_for_licensors
37
+
38
+ Considerations for the public: By using one of our public
39
+ licenses, a licensor grants the public permission to use the
40
+ licensed material under specified terms and conditions. If
41
+ the licensor's permission is not necessary for any reason--for
42
+ example, because of any applicable exception or limitation to
43
+ copyright--then that use is not regulated by the license. Our
44
+ licenses grant only permissions under copyright and certain
45
+ other rights that a licensor has authority to grant. Use of
46
+ the licensed material may still be restricted for other
47
+ reasons, including because others have copyright or other
48
+ rights in the material. A licensor may make special requests,
49
+ such as asking that all changes be marked or described.
50
+ Although not required by our licenses, you are encouraged to
51
+ respect those requests where reasonable. More considerations
52
+ for the public:
53
+ wiki.creativecommons.org/Considerations_for_licensees
54
+
55
+ =======================================================================
56
+
57
+ Creative Commons Attribution-NonCommercial 4.0 International Public
58
+ License
59
+
60
+ By exercising the Licensed Rights (defined below), You accept and agree
61
+ to be bound by the terms and conditions of this Creative Commons
62
+ Attribution-NonCommercial 4.0 International Public License ("Public
63
+ License"). To the extent this Public License may be interpreted as a
64
+ contract, You are granted the Licensed Rights in consideration of Your
65
+ acceptance of these terms and conditions, and the Licensor grants You
66
+ such rights in consideration of benefits the Licensor receives from
67
+ making the Licensed Material available under these terms and
68
+ conditions.
69
+
70
+
71
+ Section 1 -- Definitions.
72
+
73
+ a. Adapted Material means material subject to Copyright and Similar
74
+ Rights that is derived from or based upon the Licensed Material
75
+ and in which the Licensed Material is translated, altered,
76
+ arranged, transformed, or otherwise modified in a manner requiring
77
+ permission under the Copyright and Similar Rights held by the
78
+ Licensor. For purposes of this Public License, where the Licensed
79
+ Material is a musical work, performance, or sound recording,
80
+ Adapted Material is always produced where the Licensed Material is
81
+ synched in timed relation with a moving image.
82
+
83
+ b. Adapter's License means the license You apply to Your Copyright
84
+ and Similar Rights in Your contributions to Adapted Material in
85
+ accordance with the terms and conditions of this Public License.
86
+
87
+ c. Copyright and Similar Rights means copyright and/or similar rights
88
+ closely related to copyright including, without limitation,
89
+ performance, broadcast, sound recording, and Sui Generis Database
90
+ Rights, without regard to how the rights are labeled or
91
+ categorized. For purposes of this Public License, the rights
92
+ specified in Section 2(b)(1)-(2) are not Copyright and Similar
93
+ Rights.
94
+ d. Effective Technological Measures means those measures that, in the
95
+ absence of proper authority, may not be circumvented under laws
96
+ fulfilling obligations under Article 11 of the WIPO Copyright
97
+ Treaty adopted on December 20, 1996, and/or similar international
98
+ agreements.
99
+
100
+ e. Exceptions and Limitations means fair use, fair dealing, and/or
101
+ any other exception or limitation to Copyright and Similar Rights
102
+ that applies to Your use of the Licensed Material.
103
+
104
+ f. Licensed Material means the artistic or literary work, database,
105
+ or other material to which the Licensor applied this Public
106
+ License.
107
+
108
+ g. Licensed Rights means the rights granted to You subject to the
109
+ terms and conditions of this Public License, which are limited to
110
+ all Copyright and Similar Rights that apply to Your use of the
111
+ Licensed Material and that the Licensor has authority to license.
112
+
113
+ h. Licensor means the individual(s) or entity(ies) granting rights
114
+ under this Public License.
115
+
116
+ i. NonCommercial means not primarily intended for or directed towards
117
+ commercial advantage or monetary compensation. For purposes of
118
+ this Public License, the exchange of the Licensed Material for
119
+ other material subject to Copyright and Similar Rights by digital
120
+ file-sharing or similar means is NonCommercial provided there is
121
+ no payment of monetary compensation in connection with the
122
+ exchange.
123
+
124
+ j. Share means to provide material to the public by any means or
125
+ process that requires permission under the Licensed Rights, such
126
+ as reproduction, public display, public performance, distribution,
127
+ dissemination, communication, or importation, and to make material
128
+ available to the public including in ways that members of the
129
+ public may access the material from a place and at a time
130
+ individually chosen by them.
131
+
132
+ k. Sui Generis Database Rights means rights other than copyright
133
+ resulting from Directive 96/9/EC of the European Parliament and of
134
+ the Council of 11 March 1996 on the legal protection of databases,
135
+ as amended and/or succeeded, as well as other essentially
136
+ equivalent rights anywhere in the world.
137
+
138
+ l. You means the individual or entity exercising the Licensed Rights
139
+ under this Public License. Your has a corresponding meaning.
140
+
141
+
142
+ Section 2 -- Scope.
143
+
144
+ a. License grant.
145
+
146
+ 1. Subject to the terms and conditions of this Public License,
147
+ the Licensor hereby grants You a worldwide, royalty-free,
148
+ non-sublicensable, non-exclusive, irrevocable license to
149
+ exercise the Licensed Rights in the Licensed Material to:
150
+
151
+ a. reproduce and Share the Licensed Material, in whole or
152
+ in part, for NonCommercial purposes only; and
153
+
154
+ b. produce, reproduce, and Share Adapted Material for
155
+ NonCommercial purposes only.
156
+
157
+ 2. Exceptions and Limitations. For the avoidance of doubt, where
158
+ Exceptions and Limitations apply to Your use, this Public
159
+ License does not apply, and You do not need to comply with
160
+ its terms and conditions.
161
+
162
+ 3. Term. The term of this Public License is specified in Section
163
+ 6(a).
164
+
165
+ 4. Media and formats; technical modifications allowed. The
166
+ Licensor authorizes You to exercise the Licensed Rights in
167
+ all media and formats whether now known or hereafter created,
168
+ and to make technical modifications necessary to do so. The
169
+ Licensor waives and/or agrees not to assert any right or
170
+ authority to forbid You from making technical modifications
171
+ necessary to exercise the Licensed Rights, including
172
+ technical modifications necessary to circumvent Effective
173
+ Technological Measures. For purposes of this Public License,
174
+ simply making modifications authorized by this Section 2(a)
175
+ (4) never produces Adapted Material.
176
+
177
+ 5. Downstream recipients.
178
+
179
+ a. Offer from the Licensor -- Licensed Material. Every
180
+ recipient of the Licensed Material automatically
181
+ receives an offer from the Licensor to exercise the
182
+ Licensed Rights under the terms and conditions of this
183
+ Public License.
184
+
185
+ b. No downstream restrictions. You may not offer or impose
186
+ any additional or different terms or conditions on, or
187
+ apply any Effective Technological Measures to, the
188
+ Licensed Material if doing so restricts exercise of the
189
+ Licensed Rights by any recipient of the Licensed
190
+ Material.
191
+
192
+ 6. No endorsement. Nothing in this Public License constitutes or
193
+ may be construed as permission to assert or imply that You
194
+ are, or that Your use of the Licensed Material is, connected
195
+ with, or sponsored, endorsed, or granted official status by,
196
+ the Licensor or others designated to receive attribution as
197
+ provided in Section 3(a)(1)(A)(i).
198
+
199
+ b. Other rights.
200
+
201
+ 1. Moral rights, such as the right of integrity, are not
202
+ licensed under this Public License, nor are publicity,
203
+ privacy, and/or other similar personality rights; however, to
204
+ the extent possible, the Licensor waives and/or agrees not to
205
+ assert any such rights held by the Licensor to the limited
206
+ extent necessary to allow You to exercise the Licensed
207
+ Rights, but not otherwise.
208
+
209
+ 2. Patent and trademark rights are not licensed under this
210
+ Public License.
211
+
212
+ 3. To the extent possible, the Licensor waives any right to
213
+ collect royalties from You for the exercise of the Licensed
214
+ Rights, whether directly or through a collecting society
215
+ under any voluntary or waivable statutory or compulsory
216
+ licensing scheme. In all other cases the Licensor expressly
217
+ reserves any right to collect such royalties, including when
218
+ the Licensed Material is used other than for NonCommercial
219
+ purposes.
220
+
221
+
222
+ Section 3 -- License Conditions.
223
+
224
+ Your exercise of the Licensed Rights is expressly made subject to the
225
+ following conditions.
226
+
227
+ a. Attribution.
228
+
229
+ 1. If You Share the Licensed Material (including in modified
230
+ form), You must:
231
+
232
+ a. retain the following if it is supplied by the Licensor
233
+ with the Licensed Material:
234
+
235
+ i. identification of the creator(s) of the Licensed
236
+ Material and any others designated to receive
237
+ attribution, in any reasonable manner requested by
238
+ the Licensor (including by pseudonym if
239
+ designated);
240
+
241
+ ii. a copyright notice;
242
+
243
+ iii. a notice that refers to this Public License;
244
+
245
+ iv. a notice that refers to the disclaimer of
246
+ warranties;
247
+
248
+ v. a URI or hyperlink to the Licensed Material to the
249
+ extent reasonably practicable;
250
+
251
+ b. indicate if You modified the Licensed Material and
252
+ retain an indication of any previous modifications; and
253
+
254
+ c. indicate the Licensed Material is licensed under this
255
+ Public License, and include the text of, or the URI or
256
+ hyperlink to, this Public License.
257
+
258
+ 2. You may satisfy the conditions in Section 3(a)(1) in any
259
+ reasonable manner based on the medium, means, and context in
260
+ which You Share the Licensed Material. For example, it may be
261
+ reasonable to satisfy the conditions by providing a URI or
262
+ hyperlink to a resource that includes the required
263
+ information.
264
+
265
+ 3. If requested by the Licensor, You must remove any of the
266
+ information required by Section 3(a)(1)(A) to the extent
267
+ reasonably practicable.
268
+
269
+ 4. If You Share Adapted Material You produce, the Adapter's
270
+ License You apply must not prevent recipients of the Adapted
271
+ Material from complying with this Public License.
272
+
273
+
274
+ Section 4 -- Sui Generis Database Rights.
275
+
276
+ Where the Licensed Rights include Sui Generis Database Rights that
277
+ apply to Your use of the Licensed Material:
278
+
279
+ a. for the avoidance of doubt, Section 2(a)(1) grants You the right
280
+ to extract, reuse, reproduce, and Share all or a substantial
281
+ portion of the contents of the database for NonCommercial purposes
282
+ only;
283
+
284
+ b. if You include all or a substantial portion of the database
285
+ contents in a database in which You have Sui Generis Database
286
+ Rights, then the database in which You have Sui Generis Database
287
+ Rights (but not its individual contents) is Adapted Material; and
288
+
289
+ c. You must comply with the conditions in Section 3(a) if You Share
290
+ all or a substantial portion of the contents of the database.
291
+
292
+ For the avoidance of doubt, this Section 4 supplements and does not
293
+ replace Your obligations under this Public License where the Licensed
294
+ Rights include other Copyright and Similar Rights.
295
+
296
+
297
+ Section 5 -- Disclaimer of Warranties and Limitation of Liability.
298
+
299
+ a. UNLESS OTHERWISE SEPARATELY UNDERTAKEN BY THE LICENSOR, TO THE
300
+ EXTENT POSSIBLE, THE LICENSOR OFFERS THE LICENSED MATERIAL AS-IS
301
+ AND AS-AVAILABLE, AND MAKES NO REPRESENTATIONS OR WARRANTIES OF
302
+ ANY KIND CONCERNING THE LICENSED MATERIAL, WHETHER EXPRESS,
303
+ IMPLIED, STATUTORY, OR OTHER. THIS INCLUDES, WITHOUT LIMITATION,
304
+ WARRANTIES OF TITLE, MERCHANTABILITY, FITNESS FOR A PARTICULAR
305
+ PURPOSE, NON-INFRINGEMENT, ABSENCE OF LATENT OR OTHER DEFECTS,
306
+ ACCURACY, OR THE PRESENCE OR ABSENCE OF ERRORS, WHETHER OR NOT
307
+ KNOWN OR DISCOVERABLE. WHERE DISCLAIMERS OF WARRANTIES ARE NOT
308
+ ALLOWED IN FULL OR IN PART, THIS DISCLAIMER MAY NOT APPLY TO YOU.
309
+
310
+ b. TO THE EXTENT POSSIBLE, IN NO EVENT WILL THE LICENSOR BE LIABLE
311
+ TO YOU ON ANY LEGAL THEORY (INCLUDING, WITHOUT LIMITATION,
312
+ NEGLIGENCE) OR OTHERWISE FOR ANY DIRECT, SPECIAL, INDIRECT,
313
+ INCIDENTAL, CONSEQUENTIAL, PUNITIVE, EXEMPLARY, OR OTHER LOSSES,
314
+ COSTS, EXPENSES, OR DAMAGES ARISING OUT OF THIS PUBLIC LICENSE OR
315
+ USE OF THE LICENSED MATERIAL, EVEN IF THE LICENSOR HAS BEEN
316
+ ADVISED OF THE POSSIBILITY OF SUCH LOSSES, COSTS, EXPENSES, OR
317
+ DAMAGES. WHERE A LIMITATION OF LIABILITY IS NOT ALLOWED IN FULL OR
318
+ IN PART, THIS LIMITATION MAY NOT APPLY TO YOU.
319
+
320
+ c. The disclaimer of warranties and limitation of liability provided
321
+ above shall be interpreted in a manner that, to the extent
322
+ possible, most closely approximates an absolute disclaimer and
323
+ waiver of all liability.
324
+
325
+
326
+ Section 6 -- Term and Termination.
327
+
328
+ a. This Public License applies for the term of the Copyright and
329
+ Similar Rights licensed here. However, if You fail to comply with
330
+ this Public License, then Your rights under this Public License
331
+ terminate automatically.
332
+
333
+ b. Where Your right to use the Licensed Material has terminated under
334
+ Section 6(a), it reinstates:
335
+
336
+ 1. automatically as of the date the violation is cured, provided
337
+ it is cured within 30 days of Your discovery of the
338
+ violation; or
339
+
340
+ 2. upon express reinstatement by the Licensor.
341
+
342
+ For the avoidance of doubt, this Section 6(b) does not affect any
343
+ right the Licensor may have to seek remedies for Your violations
344
+ of this Public License.
345
+
346
+ c. For the avoidance of doubt, the Licensor may also offer the
347
+ Licensed Material under separate terms or conditions or stop
348
+ distributing the Licensed Material at any time; however, doing so
349
+ will not terminate this Public License.
350
+
351
+ d. Sections 1, 5, 6, 7, and 8 survive termination of this Public
352
+ License.
353
+
354
+
355
+ Section 7 -- Other Terms and Conditions.
356
+
357
+ a. The Licensor shall not be bound by any additional or different
358
+ terms or conditions communicated by You unless expressly agreed.
359
+
360
+ b. Any arrangements, understandings, or agreements regarding the
361
+ Licensed Material not stated herein are separate from and
362
+ independent of the terms and conditions of this Public License.
363
+
364
+
365
+ Section 8 -- Interpretation.
366
+
367
+ a. For the avoidance of doubt, this Public License does not, and
368
+ shall not be interpreted to, reduce, limit, restrict, or impose
369
+ conditions on any use of the Licensed Material that could lawfully
370
+ be made without permission under this Public License.
371
+
372
+ b. To the extent possible, if any provision of this Public License is
373
+ deemed unenforceable, it shall be automatically reformed to the
374
+ minimum extent necessary to make it enforceable. If the provision
375
+ cannot be reformed, it shall be severed from this Public License
376
+ without affecting the enforceability of the remaining terms and
377
+ conditions.
378
+
379
+ c. No term or condition of this Public License will be waived and no
380
+ failure to comply consented to unless expressly agreed to by the
381
+ Licensor.
382
+
383
+ d. Nothing in this Public License constitutes or may be interpreted
384
+ as a limitation upon, or waiver of, any privileges and immunities
385
+ that apply to the Licensor or You, including from the legal
386
+ processes of any jurisdiction or authority.
387
+
388
+ =======================================================================
389
+
390
+ Creative Commons is not a party to its public
391
+ licenses. Notwithstanding, Creative Commons may elect to apply one of
392
+ its public licenses to material it publishes and in those instances
393
+ will be considered the “Licensor.” The text of the Creative Commons
394
+ public licenses is dedicated to the public domain under the CC0 Public
395
+ Domain Dedication. Except for the limited purpose of indicating that
396
+ material is shared under a Creative Commons public license or as
397
+ otherwise permitted by the Creative Commons policies published at
398
+ creativecommons.org/policies, Creative Commons does not authorize the
399
+ use of the trademark "Creative Commons" or any other trademark or logo
400
+ of Creative Commons without its prior written consent including,
401
+ without limitation, in connection with any unauthorized modifications
402
+ to any of its public licenses or any other arrangements,
403
+ understandings, or agreements concerning use of licensed material. For
404
+ the avoidance of doubt, this paragraph does not form part of the
405
+ public licenses.
406
+
407
+ Creative Commons may be contacted at creativecommons.org.
408
+
LICENSE-APACHE 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.
README-Hcompany.md ADDED
@@ -0,0 +1,100 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ pipeline_tag: image-text-to-text
4
+ license: cc-by-nc-4.0
5
+ base_model: Hcompany/Holo4-27B
6
+ base_model_relation: quantized
7
+ tags:
8
+ - multimodal
9
+ - computer-use
10
+ - agent
11
+ ---
12
+
13
+ # *Holo4*-27B-NVFP4
14
+
15
+ > [!IMPORTANT]
16
+ > *Holo4* family:
17
+ >
18
+ > - Holo4-27B: Qwen3.8 dense architecture. [BF16](https://huggingface.co/Hcompany/Holo4-27B), [FP8](https://huggingface.co/Hcompany/Holo4-27B-FP8), [NVFP4](https://huggingface.co/Hcompany/Holo4-27B-NVFP4) [current], and [Q4 GGUF](https://huggingface.co/Hcompany/Holo4-27B-GGUF).
19
+ > - Holo4-35B-A3B: Qwen3.6 MoE architecture. [BF16](https://huggingface.co/Hcompany/Holo4-35B-A3B), [FP8](https://huggingface.co/Hcompany/Holo4-35B-A3B-FP8), [NVFP4](https://huggingface.co/Hcompany/Holo4-35B-A3B-NVFP4), and [Q4 GGUF](https://huggingface.co/Hcompany/Holo4-35B-A3B-GGUF).
20
+ > - Holotron4-30B-A3B: NemotronH Nano Omni architecture. [BF16](https://huggingface.co/Hcompany/Holotron4-30B-A3B) and [FP8](https://huggingface.co/Hcompany/Holotron4-30B-A3B-FP8).
21
+
22
+ [![Hugging Face: Holo4 collection](https://img.shields.io/badge/Holo4_collection-FFD21E?style=for-the-badge&logo=huggingface&logoColor=000)](https://hf.co/collections/Hcompany/holo4)
23
+ [![Blog post](https://img.shields.io/badge/Blog_post-181717?style=for-the-badge)](https://hcompany.ai/newsroom/holo4)
24
+ [![H Models API](https://img.shields.io/badge/H_Models_API-181717?style=for-the-badge)](https://hub.hcompany.ai/models-api/introduction)
25
+ [![Agent trajectories](https://img.shields.io/badge/Agent_trajectories-181717?style=for-the-badge)](https://trajectories.hcompany.ai/)
26
+
27
+ ## Model summary
28
+
29
+ *Holo4*-27B-NVFP4 is a vision-language model (VLM) for Computer Use, built on [Qwen3.8-27B](https://huggingface.co/Qwen/Qwen3.8-27B) and developed by [H Company](https://hcompany.ai/). Used with the [hai-agents harness](https://github.com/hcompai/hai-agents-python), it can send screenshots and tool results to the model, then execute its requested clicks, typing, code, and tool calls.
30
+
31
+ <table width="100%">
32
+ <thead>
33
+ <tr style="border-bottom: 2px solid rgb(230, 126, 34);"><th style="background-color: rgb(255, 248, 239); color: #1f2937;">Specification</th><th style="background-color: rgb(255, 248, 239); color: #1f2937;">Value</th></tr>
34
+ </thead>
35
+ <tbody>
36
+ <tr><td style="background-color: white; color: #1f2937;">Parameters</td><td style="background-color: white; color: #1f2937;">27B dense</td></tr>
37
+ <tr><td style="background-color: white; color: #1f2937;">Maximum context length in config</td><td style="background-color: white; color: #1f2937;">262,144 tokens</td></tr>
38
+ <tr><td style="background-color: white; color: #1f2937;">Model type</td><td style="background-color: white; color: #1f2937;">Visual Language Model (VLM)</td></tr>
39
+ <tr><td style="background-color: white; color: #1f2937;">Architecture</td><td style="background-color: white; color: #1f2937;">Qwen3.8 dense</td></tr>
40
+ <tr><td style="background-color: white; color: #1f2937;">Interfaces</td><td style="background-color: white; color: #1f2937;">Graphical interface, code, and tool calls</td></tr>
41
+ <tr><td style="background-color: white; color: #1f2937;">Target environments</td><td style="background-color: white; color: #1f2937;">Web, desktop, and mobile</td></tr>
42
+ </tbody>
43
+ </table>
44
+
45
+ <video src="https://assets.hcompanyprod.fr/holo4/eiffel-holo4.mp4" poster="https://assets.hcompanyprod.fr/holo4/eiffel-holo4-poster.jpg" controls muted loop playsinline width="100%"></video>
46
+
47
+ *Holo4* uses FreeCAD to build a replica of the Eiffel Tower. More examples in the [blog post](https://hcompany.ai/newsroom/holo4).
48
+
49
+ <details>
50
+ <summary>Prompt</summary>
51
+
52
+ <blockquote>
53
+ <p>Build a 3D Eiffel Tower in FreeCAD at a scale of 1 mm per metre. Center it on the origin and align it with the X and Y axes. Its plan must stay square at every height. The distance from the center to each corner is 62.5 mm at ground level, 32.5 mm at height 57, 17.5 mm at height 115, and 9.35 mm at height 276. Connect these widths with a smooth curve that narrows quickly near the base and more slowly near the top.</p>
54
+
55
+ <p>Make four separate, identical square legs, one in each quadrant. Their outer corners follow that curve, and each leg narrows from 14 mm across at ground level to 4 mm at height 276. Leave the space between the legs open. Add centered square platforms measuring 72 mm by 72 mm by 4 mm at height 57, 40 mm by 40 mm by 3 mm at height 115, and 22 mm by 22 mm by 3 mm at height 276. Add a square mast from height 276 to 324, tapering from 8 mm across to 2 mm across. Make every component a closed solid with nonzero volume, without filling the space between the legs.</p>
56
+ </blockquote>
57
+
58
+ </details>
59
+
60
+ ## Performance
61
+
62
+ *Holo4* models improve significantly over their Qwen base models. On OSWorld, *Holo4*-27B scores 85.2% at $0.08 per task. We also evaluate them on Agentic Task Factory, a set of held-out business workflows across web, desktop, and MCP tools.
63
+
64
+ ### Benchmark results
65
+
66
+ ![Holo4 benchmark comparison table](https://assets.hcompanyprod.fr/holo4/hf/holo4-benchmark-table.png)
67
+
68
+ Read the [blog post](https://hcompany.ai/newsroom/holo4) for more tables about the benchmark results.
69
+
70
+ ### Pareto plots
71
+
72
+ These plots compare benchmark scores with cost per task. On OSWorld 2.0, *Holo4*-27B scores 61.7% at $1.22 per task, and *Holo4*-35B-A3B scores 30.9% at $0.61 per task.
73
+
74
+ ![Cost versus score on OSWorld 2.0](https://assets.hcompanyprod.fr/holo4/hf/holo4-cost-osworld2.png)
75
+
76
+ On AutomationBench, *Holo4*-27B scores 45.4% at $0.05 per task, and *Holo4*-35B-A3B scores 34.5% at $0.02 per task.
77
+
78
+ ![Cost versus score on AutomationBench](https://assets.hcompanyprod.fr/holo4/hf/holo4-cost-automationbench.png)
79
+
80
+ ### Open-source evaluation traces
81
+
82
+ For transparency, we share all agent trajectories in the open-source dataset at [Hcompany/trajectories](https://huggingface.co/datasets/Hcompany/trajectories).
83
+
84
+ ## Usage
85
+
86
+ The harness sends screenshots and tool results to *Holo4*, executes the model's requested actions, and sends the results back. It can give the model access to application tools and code execution.
87
+
88
+ Refer to the documentation for more details about:
89
+
90
+ - [Function calling](https://hub.hcompany.ai/models-api/build-an-agent/function-calling)
91
+ - [Element localization](https://hub.hcompany.ai/models-api/element-localization)
92
+ - [Document OCR](https://hub.hcompany.ai/models-api/document-ocr)
93
+
94
+ ## Training
95
+
96
+ ![Holo4 training pipeline](https://assets.hcompanyprod.fr/holo4/hf/holo4-training.png)
97
+
98
+ ## License
99
+
100
+ The model weights are available under the Non-commercial ([CC BY-NC 4.0](https://creativecommons.org/licenses/by-nc/4.0/)) license. This model is built on [Qwen3.8-27B](https://huggingface.co/Qwen/Qwen3.8-27B), which Alibaba Cloud releases under the Apache License 2.0. The repository includes both licenses.
README.md ADDED
@@ -0,0 +1,121 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ pipeline_tag: image-text-to-text
4
+ license: cc-by-nc-4.0
5
+ base_model: Hcompany/Holo4-27B
6
+ base_model_relation: quantized
7
+ tags:
8
+ - multimodal
9
+ - computer-use
10
+ - agent
11
+ - nvfp4
12
+ - compressed-tensors
13
+ - mtp
14
+ - speculative-decoding
15
+ - vllm
16
+ ---
17
+
18
+ # Holo4-27B-NVFP4-MTP
19
+
20
+ Community packaging by RubyWitch of [H Company's Holo4-27B-NVFP4](https://huggingface.co/Hcompany/Holo4-27B-NVFP4), with the compatible BF16 multi-token prediction (MTP) head from [Qwen/Qwen3.8-27B](https://huggingface.co/Qwen/Qwen3.8-27B) included for speculative decoding in vLLM.
21
+
22
+ **The original Holo4 target weights are unchanged.** This repository adds the donor MTP head and the metadata needed to load it. No additional training or requantization was performed. The donor head comes from Qwen, rather than from a Holo4-specific MTP fine-tune. This is an unofficial community derivative, not an H Company or Qwen release.
23
+
24
+ ## Provenance and changes
25
+
26
+ | Component | Source | Pinned revision |
27
+ | --- | --- | --- |
28
+ | NVFP4 target weights, tokenizer, processor and chat template | [Hcompany/Holo4-27B-NVFP4](https://huggingface.co/Hcompany/Holo4-27B-NVFP4/tree/6a02909667ba91dcd45e01be13c4d49992112510) | `6a02909667ba91dcd45e01be13c4d49992112510` |
29
+ | BF16 MTP head | [Qwen/Qwen3.8-27B](https://huggingface.co/Qwen/Qwen3.8-27B/tree/1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0) | `1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0` |
30
+
31
+ The upstream NVFP4 checkpoint declares one MTP layer but does not contain its weights. This package:
32
+
33
+ - Preserves both original Holo4 weight shards byte for byte.
34
+ - Adds `model-mtp.safetensors`, containing exactly 15 `mtp.*` BF16 tensors extracted from the donor's final shard. Tensor data occupies 849,398,784 bytes; the complete shard is 849,400,424 bytes.
35
+ - Excludes the donor's separate `lm_head.weight`. Target and draft use Holo4's embeddings and output head.
36
+ - Adds the MTP tensors to `model.safetensors.index.json` and updates its total size.
37
+ - Adds `re:^mtp\.` to the compressed-tensors quantization ignore list, keeping the draft head in BF16.
38
+ - Keeps the target's 64-layer count unchanged; the MTP layer is separate.
39
+ - Preserves the tokenizer, processor, generation configuration, chat template and original quantization recipe.
40
+
41
+ [crystal-mtp-provenance.json](./crystal-mtp-provenance.json) records immutable source identities, source file hashes, per-tensor hashes and hashes of the modified configuration, index and new shard. [SHA256SUMS](./SHA256SUMS) covers this package's files. [README-Hcompany.md](./README-Hcompany.md) preserves the original model card, including its attribution and links.
42
+
43
+ ## Running with vLLM
44
+
45
+ Tested on 2026-10-06 with a native NVIDIA GB10 workstation and vLLM source revision [`52358e6e192aeae73dd3764046c98fc4b156ea83`](https://github.com/vllm-project/vllm/tree/52358e6e192aeae73dd3764046c98fc4b156ea83). Use a build supporting this architecture, compressed-tensors NVFP4 and Qwen MTP; older vLLM versions may lack the required functionality or flags.
46
+
47
+ The following reproduces the tested **128K context, two concurrent requests, three speculative tokens** profile. Its memory settings are specific to the GB10 workstation and must be adjusted for other hardware and workloads.
48
+
49
+ ```bash
50
+ vllm serve RubyWitch/Holo4-27B-NVFP4-MTP \
51
+ --served-model-name Holo4-27B-NVFP4-MTP \
52
+ --host 127.0.0.1 --port 30000 \
53
+ --dtype bfloat16 \
54
+ --quantization compressed-tensors \
55
+ --max-model-len 131072 \
56
+ --max-num-seqs 2 \
57
+ --gpu-memory-utilization 0.55 \
58
+ --kv-cache-memory-bytes 30064771072 \
59
+ --kv-cache-dtype bfloat16 \
60
+ --mamba-ssm-cache-dtype float32 \
61
+ --enable-prefix-caching \
62
+ --prefix-cache-retention-interval 800 \
63
+ --mamba-cache-mode align \
64
+ --enable-chunked-prefill \
65
+ --max-num-batched-tokens 8192 \
66
+ --async-scheduling \
67
+ --speculative-config '{"method":"mtp","num_speculative_tokens":3}' \
68
+ --chat-template-content-format openai \
69
+ --enable-auto-tool-choice \
70
+ --tool-call-parser qwen3_coder \
71
+ --reasoning-parser qwen3 \
72
+ --limit-mm-per-prompt '{"image":5,"video":0}' \
73
+ --mm-processor-cache-type shm \
74
+ --mm-processor-cache-gb 4
75
+ ```
76
+
77
+ The repository includes one MTP layer. `num_speculative_tokens: 3` reuses that head for three drafting steps; it does not require three separate heads.
78
+
79
+ On the tested GB10 runtime, vLLM automatically selected Marlin for the NVFP4 W4A16 layers, retained its usual backend selection for other FP8 layers, and used FlashAttention 2 for BF16 attention. No global Marlin override, eager-mode fallback or runtime source patch was required. This does not establish the best kernel choice on other GPUs.
80
+
81
+ The explicit KV-cache allocation is **28 GiB**, not the total memory requirement. With this flag, `gpu_memory_utilization` is not a 55% total-allocation cap in the tested runtime. Weights, activations, Mamba state, processor caches, compilation and other runtime allocations need additional memory. The 800-token prefix-retention interval matches the block size selected for this particular MTP depth and cache precision; review it if changing those settings. The model configuration retains the upstream context limit, but this package was qualified at 131,072 tokens.
82
+
83
+ Keep sufficient disk space available for CUDA/FlashInfer compilation scratch. On the test workstation, compilation initially exhausted a quota-limited `/tmp`; selecting a larger writable `TMPDIR` resolved that startup failure.
84
+
85
+ Example text request:
86
+
87
+ ```bash
88
+ curl http://127.0.0.1:30000/v1/chat/completions \
89
+ -H 'Content-Type: application/json' \
90
+ -d '{
91
+ "model": "Holo4-27B-NVFP4-MTP",
92
+ "messages": [{"role": "user", "content": "Explain speculative decoding briefly."}],
93
+ "max_tokens": 256,
94
+ "chat_template_kwargs": {"enable_thinking": false}
95
+ }'
96
+ ```
97
+
98
+ See H Company's documentation for [function calling](https://hub.hcompany.ai/models-api/build-an-agent/function-calling) and [element localization](https://hub.hcompany.ai/models-api/element-localization).
99
+
100
+ ## Local qualification
101
+
102
+ Functional checks passed for text generation, greedy and sampled image recognition, reasoning output, automatic tool calls and tool-result continuation, constrained-JSON grounding, and prefix-cache reuse.
103
+
104
+ Two simultaneous requests each used 130,944 input tokens plus 128 output tokens, recovered their distinct test codes, and completed with zero cache preemptions. Peak KV-cache use was 83.27% of the 28 GiB allocation. That stress test took about 391 seconds overall; it is not a 128K-context decode-speed benchmark.
105
+
106
+ Short-context measurements on the shared GB10 workstation used 512 generated tokens per prompt. Sequential decode rates exclude time to first token; the concurrent aggregate includes full request time.
107
+
108
+ | Profile | Code tokens/s | Explanatory text tokens/s | Screenshot response tokens/s | Two-request aggregate tokens/s |
109
+ | --- | ---: | ---: | ---: | ---: |
110
+ | No MTP, earlier automatic cache budget | 9.56 | 9.64 | 9.62 | 18.77 |
111
+ | MTP 3, final fixed 28 GiB cache | 20.88 | 16.80 | 18.66 | 37.94 |
112
+
113
+ These are individual local runs with different cache budgets, not repeated, isolated performance trials or a general speed guarantee. Comparing three and five speculative tokens at the same earlier cache budget favored three for mixed use: five slightly improved code but reduced explanatory-text throughput. Acceptance and speed depend on prompts, sampling, context length, hardware and runtime version.
114
+
115
+ No independent quality benchmark of this composite has been performed. Upstream Holo4 benchmark claims belong to the upstream release and should not be treated as new evaluations of this package. The compatibility checks above establish serving functionality, not comprehensive accuracy equivalence across workloads.
116
+
117
+ ## License and attribution
118
+
119
+ Holo4 was developed by **H Company** and builds on **Qwen3.8-27B from Alibaba Cloud / the Qwen team**. The NVFP4 target weights and related upstream files are redistributed with H Company's **CC BY-NC 4.0** license in [LICENSE](./LICENSE). The Qwen MTP tensors originate from the **Apache-2.0** licensed Qwen model; its license is retained in [LICENSE-APACHE](./LICENSE-APACHE).
120
+
121
+ This community packaging does not remove Holo4's noncommercial restriction. The modifications are listed above; original attribution and license texts are retained. Consult the [upstream model card](https://huggingface.co/Hcompany/Holo4-27B) for the model's intended uses and limitations.
SHA256SUMS ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 34448b82c17d60fec9b65b1f093c115ddbaadc04beb1b0140b6bfed2e012a930 .gitattributes
2
+ 41003d4a74749c0220e33dd415042164b5a1093ed401f36277234f772d22d3d0 LICENSE
3
+ bbedc3fda3305820b977265f01b8619d87570a6739de3a5582c3464840f1e57a LICENSE-APACHE
4
+ 5311535fee699a112d424787f44240549253b6d035f80872dcb8c7d5bc0fb8c5 README-Hcompany.md
5
+ 57f0768b7988d5663094cebb5a171823a78359fcd6c5d82e19a99131ee073d35 README.md
6
+ c3cf9e34abf4f9e36c2d72165aa9c132d3e2a725b6c2586aaa3a8af9d7a81041 chat_template.jinja
7
+ 8fe4a28d2437e47e12246b1a5e4415f848e4dc89fb86ccfcd1cd32a8ba09f52f config.json
8
+ 0c4f0826f79a8be44c999afcfc380ae86ab5f318dab27902e255d86c221cfae7 crystal-mtp-provenance.json
9
+ 56147698c439ec2da22f5820fcf76fa14788b6c0ebdd2b2e073c6cddfdd1ea30 generation_config.json
10
+ 2cdf9e105d274204894a086d44dc805d1a4b428b9e9d8220c58b5a21972ce54f model-00001-of-00002.safetensors
11
+ 229ba26491580906dfdf595c9ad4a73b1fd258895fce67f3cde3f2ddb595ff03 model-00002-of-00002.safetensors
12
+ 90fa0e3eed5a647c035c6df9ecabc416c0f8d573ff84ac12485b085f00a7cdf2 model-mtp.safetensors
13
+ 4a701c716a3b5616eafba7402e70907ef8f7e0266d929d7bd14417d58cb0dc98 model.safetensors.index.json
14
+ d89ef49ce9cd37fbf510158e13c1ef063d9286411c1ec9049932dbe0487143b1 processor_config.json
15
+ f79d08cba8557097a26c6fc4713f9d214b2ba7850105d98d77f9fc9ed232d22f recipe.yaml
16
+ 06b9509352d2af50381ab2247e083b80d32d5c0aba91c272ca9ff729b6a0e523 tokenizer.json
17
+ 792fa3f0cb88b111e54ef3134c873531008c4df471d108da17903426e308aa7b tokenizer_config.json
chat_template.jinja ADDED
@@ -0,0 +1,170 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ {%- set reasoning_instructions = '' %}
46
+ {%- if enable_thinking is undefined or enable_thinking is true %}
47
+ {%- set resolved_reasoning_effort = reasoning_effort|default('xhigh') %}
48
+ {%- if resolved_reasoning_effort not in ('xhigh', 'medium', 'low') %}
49
+ {{- raise_exception('Unexpected reasoning effort ' ~ reasoning_effort ~ '. Supported types are xhigh (default), medium, and low.') }}
50
+ {%- endif %}
51
+ {%- if resolved_reasoning_effort == 'xhigh' %}
52
+ {%- set reasoning_instructions = 'Reasoning effort is set to xhigh. Please think carefully through the task, validate key assumptions, consider plausible alternatives, and prioritize correctness, consistency, and clarity in the final answer.' %}
53
+ {%- elif resolved_reasoning_effort == 'low' %}
54
+ {%- set reasoning_instructions = 'Reasoning effort is set to low. Keep your thinking brief and focused, moving directly to the conclusion without unnecessary elaboration.' %}
55
+ {%- endif %}
56
+ {%- endif %}
57
+ {%- if tools and tools is iterable and tools is not mapping %}
58
+ {{- '<|im_start|>system\n' }}
59
+ {%- if reasoning_instructions %}
60
+ {{- reasoning_instructions + '\n\n' }}
61
+ {%- endif %}
62
+ {{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
63
+ {%- for tool in tools %}
64
+ {{- "\n" }}
65
+ {{- tool | tojson }}
66
+ {%- endfor %}
67
+ {{- "\n</tools>" }}
68
+ {{- '\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>' }}
69
+ {%- if messages[0].role == 'system' %}
70
+ {%- set content = render_content(messages[0].content, false, true)|trim %}
71
+ {%- if content %}
72
+ {{- '\n\n' + content }}
73
+ {%- endif %}
74
+ {%- endif %}
75
+ {{- '<|im_end|>\n' }}
76
+ {%- else %}
77
+ {%- if messages[0].role == 'system' %}
78
+ {%- set content = render_content(messages[0].content, false, true)|trim %}
79
+ {%- if content %}
80
+ {{- '<|im_start|>system\n' + (reasoning_instructions + '\n\n' if reasoning_instructions else '') + content + '<|im_end|>\n' }}
81
+ {%- elif reasoning_instructions %}
82
+ {{- '<|im_start|>system\n' + reasoning_instructions + '<|im_end|>\n' }}
83
+ {%- endif %}
84
+ {%- elif reasoning_instructions %}
85
+ {{- '<|im_start|>system\n' + reasoning_instructions + '<|im_end|>\n' }}
86
+ {%- endif %}
87
+ {%- endif %}
88
+ {%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
89
+ {%- for message in messages[::-1] %}
90
+ {%- set index = (messages|length - 1) - loop.index0 %}
91
+ {%- if ns.multi_step_tool and message.role == "user" %}
92
+ {%- set content = render_content(message.content, false)|trim %}
93
+ {%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
94
+ {%- set ns.multi_step_tool = false %}
95
+ {%- set ns.last_query_index = index %}
96
+ {%- endif %}
97
+ {%- endif %}
98
+ {%- endfor %}
99
+ {%- if ns.multi_step_tool %}
100
+ {{- raise_exception('No user query found in messages.') }}
101
+ {%- endif %}
102
+ {%- for message in messages %}
103
+ {%- set content = render_content(message.content, true)|trim %}
104
+ {%- if message.role == "system" %}
105
+ {%- if not loop.first %}
106
+ {{- raise_exception('System message must be at the beginning.') }}
107
+ {%- endif %}
108
+ {%- elif message.role == "user" %}
109
+ {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
110
+ {%- elif message.role == "assistant" %}
111
+ {%- set reasoning_content = '' %}
112
+ {%- if message.reasoning_content is string %}
113
+ {%- set reasoning_content = message.reasoning_content %}
114
+ {%- endif %}
115
+ {%- set reasoning_content = reasoning_content|trim %}
116
+ {%- if preserve_thinking is undefined or preserve_thinking is true or loop.index0 > ns.last_query_index %}
117
+ {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
118
+ {%- else %}
119
+ {{- '<|im_start|>' + message.role + '\n' + content }}
120
+ {%- endif %}
121
+ {%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
122
+ {%- for tool_call in message.tool_calls %}
123
+ {%- if tool_call.function is defined %}
124
+ {%- set tool_call = tool_call.function %}
125
+ {%- endif %}
126
+ {%- if loop.first %}
127
+ {%- if content|trim %}
128
+ {{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
129
+ {%- else %}
130
+ {{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
131
+ {%- endif %}
132
+ {%- else %}
133
+ {{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
134
+ {%- endif %}
135
+ {%- if tool_call.arguments is defined and tool_call.arguments != '' %}
136
+ {%- for args_name, args_value in tool_call.arguments|items %}
137
+ {{- '<parameter=' + args_name + '>\n' }}
138
+ {%- set args_value = args_value | string if args_value is string else args_value | tojson | safe %}
139
+ {{- args_value }}
140
+ {{- '\n</parameter>\n' }}
141
+ {%- endfor %}
142
+ {%- endif %}
143
+ {{- '</function>\n</tool_call>' }}
144
+ {%- endfor %}
145
+ {%- endif %}
146
+ {{- '<|im_end|>\n' }}
147
+ {%- elif message.role == "tool" %}
148
+ {%- if loop.previtem and loop.previtem.role != "tool" %}
149
+ {{- '<|im_start|>user' }}
150
+ {%- endif %}
151
+ {{- '\n<tool_response>\n' }}
152
+ {{- content }}
153
+ {{- '\n</tool_response>' }}
154
+ {%- if not loop.last and loop.nextitem.role != "tool" %}
155
+ {{- '<|im_end|>\n' }}
156
+ {%- elif loop.last %}
157
+ {{- '<|im_end|>\n' }}
158
+ {%- endif %}
159
+ {%- else %}
160
+ {{- raise_exception('Unexpected message role.') }}
161
+ {%- endif %}
162
+ {%- endfor %}
163
+ {%- if add_generation_prompt %}
164
+ {{- '<|im_start|>assistant\n' }}
165
+ {%- if enable_thinking is defined and enable_thinking is false %}
166
+ {{- '<think>\n\n</think>\n\n' }}
167
+ {%- else %}
168
+ {{- '<think>\n' }}
169
+ {%- endif %}
170
+ {%- endif %}
config.json ADDED
@@ -0,0 +1,537 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3_5ForConditionalGeneration"
4
+ ],
5
+ "dtype": "bfloat16",
6
+ "head_dim": 256,
7
+ "image_token_id": 248056,
8
+ "language_model_only": false,
9
+ "model_type": "qwen3_5",
10
+ "num_attention_heads": 24,
11
+ "num_key_value_heads": 4,
12
+ "quantization_config": {
13
+ "config_groups": {
14
+ "group_0": {
15
+ "format": "float-quantized",
16
+ "input_activations": {
17
+ "actorder": null,
18
+ "block_structure": null,
19
+ "dynamic": false,
20
+ "group_size": null,
21
+ "num_bits": 8,
22
+ "observer": "static_minmax",
23
+ "observer_kwargs": {},
24
+ "scale_dtype": null,
25
+ "strategy": "tensor",
26
+ "symmetric": true,
27
+ "type": "float",
28
+ "zp_dtype": null
29
+ },
30
+ "output_activations": null,
31
+ "targets": [
32
+ "re:.*self_attn\\.(q|k|v|o)_proj$",
33
+ "re:.*linear_attn\\.(in_proj_qkv|in_proj_z|out_proj)$"
34
+ ],
35
+ "weights": {
36
+ "actorder": null,
37
+ "block_structure": null,
38
+ "dynamic": false,
39
+ "group_size": null,
40
+ "num_bits": 8,
41
+ "observer": "memoryless_minmax",
42
+ "observer_kwargs": {},
43
+ "scale_dtype": null,
44
+ "strategy": "tensor",
45
+ "symmetric": true,
46
+ "type": "float",
47
+ "zp_dtype": null
48
+ }
49
+ },
50
+ "group_1": {
51
+ "format": "nvfp4-pack-quantized",
52
+ "input_activations": null,
53
+ "output_activations": null,
54
+ "targets": [
55
+ "re:.*mlp\\.experts\\.\\d+\\.(gate|up|down)_proj$",
56
+ "re:.*mlp\\.shared_expert\\.(gate|up|down)_proj$",
57
+ "re:.*mlp\\.(gate|up|down)_proj$"
58
+ ],
59
+ "weights": {
60
+ "actorder": null,
61
+ "block_structure": null,
62
+ "dynamic": false,
63
+ "group_size": 16,
64
+ "num_bits": 4,
65
+ "observer": "memoryless_minmax",
66
+ "observer_kwargs": {},
67
+ "scale_dtype": "torch.float8_e4m3fn",
68
+ "strategy": "tensor_group",
69
+ "symmetric": true,
70
+ "type": "float",
71
+ "zp_dtype": null
72
+ }
73
+ }
74
+ },
75
+ "format": "mixed-precision",
76
+ "global_compression_ratio": null,
77
+ "ignore": [
78
+ "model.visual.blocks.0.attn.qkv",
79
+ "model.visual.blocks.0.attn.proj",
80
+ "model.visual.blocks.0.mlp.linear_fc1",
81
+ "model.visual.blocks.0.mlp.linear_fc2",
82
+ "model.visual.blocks.1.attn.qkv",
83
+ "model.visual.blocks.1.attn.proj",
84
+ "model.visual.blocks.1.mlp.linear_fc1",
85
+ "model.visual.blocks.1.mlp.linear_fc2",
86
+ "model.visual.blocks.2.attn.qkv",
87
+ "model.visual.blocks.2.attn.proj",
88
+ "model.visual.blocks.2.mlp.linear_fc1",
89
+ "model.visual.blocks.2.mlp.linear_fc2",
90
+ "model.visual.blocks.3.attn.qkv",
91
+ "model.visual.blocks.3.attn.proj",
92
+ "model.visual.blocks.3.mlp.linear_fc1",
93
+ "model.visual.blocks.3.mlp.linear_fc2",
94
+ "model.visual.blocks.4.attn.qkv",
95
+ "model.visual.blocks.4.attn.proj",
96
+ "model.visual.blocks.4.mlp.linear_fc1",
97
+ "model.visual.blocks.4.mlp.linear_fc2",
98
+ "model.visual.blocks.5.attn.qkv",
99
+ "model.visual.blocks.5.attn.proj",
100
+ "model.visual.blocks.5.mlp.linear_fc1",
101
+ "model.visual.blocks.5.mlp.linear_fc2",
102
+ "model.visual.blocks.6.attn.qkv",
103
+ "model.visual.blocks.6.attn.proj",
104
+ "model.visual.blocks.6.mlp.linear_fc1",
105
+ "model.visual.blocks.6.mlp.linear_fc2",
106
+ "model.visual.blocks.7.attn.qkv",
107
+ "model.visual.blocks.7.attn.proj",
108
+ "model.visual.blocks.7.mlp.linear_fc1",
109
+ "model.visual.blocks.7.mlp.linear_fc2",
110
+ "model.visual.blocks.8.attn.qkv",
111
+ "model.visual.blocks.8.attn.proj",
112
+ "model.visual.blocks.8.mlp.linear_fc1",
113
+ "model.visual.blocks.8.mlp.linear_fc2",
114
+ "model.visual.blocks.9.attn.qkv",
115
+ "model.visual.blocks.9.attn.proj",
116
+ "model.visual.blocks.9.mlp.linear_fc1",
117
+ "model.visual.blocks.9.mlp.linear_fc2",
118
+ "model.visual.blocks.10.attn.qkv",
119
+ "model.visual.blocks.10.attn.proj",
120
+ "model.visual.blocks.10.mlp.linear_fc1",
121
+ "model.visual.blocks.10.mlp.linear_fc2",
122
+ "model.visual.blocks.11.attn.qkv",
123
+ "model.visual.blocks.11.attn.proj",
124
+ "model.visual.blocks.11.mlp.linear_fc1",
125
+ "model.visual.blocks.11.mlp.linear_fc2",
126
+ "model.visual.blocks.12.attn.qkv",
127
+ "model.visual.blocks.12.attn.proj",
128
+ "model.visual.blocks.12.mlp.linear_fc1",
129
+ "model.visual.blocks.12.mlp.linear_fc2",
130
+ "model.visual.blocks.13.attn.qkv",
131
+ "model.visual.blocks.13.attn.proj",
132
+ "model.visual.blocks.13.mlp.linear_fc1",
133
+ "model.visual.blocks.13.mlp.linear_fc2",
134
+ "model.visual.blocks.14.attn.qkv",
135
+ "model.visual.blocks.14.attn.proj",
136
+ "model.visual.blocks.14.mlp.linear_fc1",
137
+ "model.visual.blocks.14.mlp.linear_fc2",
138
+ "model.visual.blocks.15.attn.qkv",
139
+ "model.visual.blocks.15.attn.proj",
140
+ "model.visual.blocks.15.mlp.linear_fc1",
141
+ "model.visual.blocks.15.mlp.linear_fc2",
142
+ "model.visual.blocks.16.attn.qkv",
143
+ "model.visual.blocks.16.attn.proj",
144
+ "model.visual.blocks.16.mlp.linear_fc1",
145
+ "model.visual.blocks.16.mlp.linear_fc2",
146
+ "model.visual.blocks.17.attn.qkv",
147
+ "model.visual.blocks.17.attn.proj",
148
+ "model.visual.blocks.17.mlp.linear_fc1",
149
+ "model.visual.blocks.17.mlp.linear_fc2",
150
+ "model.visual.blocks.18.attn.qkv",
151
+ "model.visual.blocks.18.attn.proj",
152
+ "model.visual.blocks.18.mlp.linear_fc1",
153
+ "model.visual.blocks.18.mlp.linear_fc2",
154
+ "model.visual.blocks.19.attn.qkv",
155
+ "model.visual.blocks.19.attn.proj",
156
+ "model.visual.blocks.19.mlp.linear_fc1",
157
+ "model.visual.blocks.19.mlp.linear_fc2",
158
+ "model.visual.blocks.20.attn.qkv",
159
+ "model.visual.blocks.20.attn.proj",
160
+ "model.visual.blocks.20.mlp.linear_fc1",
161
+ "model.visual.blocks.20.mlp.linear_fc2",
162
+ "model.visual.blocks.21.attn.qkv",
163
+ "model.visual.blocks.21.attn.proj",
164
+ "model.visual.blocks.21.mlp.linear_fc1",
165
+ "model.visual.blocks.21.mlp.linear_fc2",
166
+ "model.visual.blocks.22.attn.qkv",
167
+ "model.visual.blocks.22.attn.proj",
168
+ "model.visual.blocks.22.mlp.linear_fc1",
169
+ "model.visual.blocks.22.mlp.linear_fc2",
170
+ "model.visual.blocks.23.attn.qkv",
171
+ "model.visual.blocks.23.attn.proj",
172
+ "model.visual.blocks.23.mlp.linear_fc1",
173
+ "model.visual.blocks.23.mlp.linear_fc2",
174
+ "model.visual.blocks.24.attn.qkv",
175
+ "model.visual.blocks.24.attn.proj",
176
+ "model.visual.blocks.24.mlp.linear_fc1",
177
+ "model.visual.blocks.24.mlp.linear_fc2",
178
+ "model.visual.blocks.25.attn.qkv",
179
+ "model.visual.blocks.25.attn.proj",
180
+ "model.visual.blocks.25.mlp.linear_fc1",
181
+ "model.visual.blocks.25.mlp.linear_fc2",
182
+ "model.visual.blocks.26.attn.qkv",
183
+ "model.visual.blocks.26.attn.proj",
184
+ "model.visual.blocks.26.mlp.linear_fc1",
185
+ "model.visual.blocks.26.mlp.linear_fc2",
186
+ "model.visual.merger.linear_fc1",
187
+ "model.visual.merger.linear_fc2",
188
+ "model.language_model.layers.0.linear_attn",
189
+ "model.language_model.layers.0.linear_attn.norm",
190
+ "model.language_model.layers.0.linear_attn.in_proj_b",
191
+ "model.language_model.layers.0.linear_attn.in_proj_a",
192
+ "model.language_model.layers.1.linear_attn",
193
+ "model.language_model.layers.1.linear_attn.norm",
194
+ "model.language_model.layers.1.linear_attn.in_proj_b",
195
+ "model.language_model.layers.1.linear_attn.in_proj_a",
196
+ "model.language_model.layers.2.linear_attn",
197
+ "model.language_model.layers.2.linear_attn.norm",
198
+ "model.language_model.layers.2.linear_attn.in_proj_b",
199
+ "model.language_model.layers.2.linear_attn.in_proj_a",
200
+ "model.language_model.layers.4.linear_attn",
201
+ "model.language_model.layers.4.linear_attn.norm",
202
+ "model.language_model.layers.4.linear_attn.in_proj_b",
203
+ "model.language_model.layers.4.linear_attn.in_proj_a",
204
+ "model.language_model.layers.5.linear_attn",
205
+ "model.language_model.layers.5.linear_attn.norm",
206
+ "model.language_model.layers.5.linear_attn.in_proj_b",
207
+ "model.language_model.layers.5.linear_attn.in_proj_a",
208
+ "model.language_model.layers.6.linear_attn",
209
+ "model.language_model.layers.6.linear_attn.norm",
210
+ "model.language_model.layers.6.linear_attn.in_proj_b",
211
+ "model.language_model.layers.6.linear_attn.in_proj_a",
212
+ "model.language_model.layers.8.linear_attn",
213
+ "model.language_model.layers.8.linear_attn.norm",
214
+ "model.language_model.layers.8.linear_attn.in_proj_b",
215
+ "model.language_model.layers.8.linear_attn.in_proj_a",
216
+ "model.language_model.layers.9.linear_attn",
217
+ "model.language_model.layers.9.linear_attn.norm",
218
+ "model.language_model.layers.9.linear_attn.in_proj_b",
219
+ "model.language_model.layers.9.linear_attn.in_proj_a",
220
+ "model.language_model.layers.10.linear_attn",
221
+ "model.language_model.layers.10.linear_attn.norm",
222
+ "model.language_model.layers.10.linear_attn.in_proj_b",
223
+ "model.language_model.layers.10.linear_attn.in_proj_a",
224
+ "model.language_model.layers.12.linear_attn",
225
+ "model.language_model.layers.12.linear_attn.norm",
226
+ "model.language_model.layers.12.linear_attn.in_proj_b",
227
+ "model.language_model.layers.12.linear_attn.in_proj_a",
228
+ "model.language_model.layers.13.linear_attn",
229
+ "model.language_model.layers.13.linear_attn.norm",
230
+ "model.language_model.layers.13.linear_attn.in_proj_b",
231
+ "model.language_model.layers.13.linear_attn.in_proj_a",
232
+ "model.language_model.layers.14.linear_attn",
233
+ "model.language_model.layers.14.linear_attn.norm",
234
+ "model.language_model.layers.14.linear_attn.in_proj_b",
235
+ "model.language_model.layers.14.linear_attn.in_proj_a",
236
+ "model.language_model.layers.16.linear_attn",
237
+ "model.language_model.layers.16.linear_attn.norm",
238
+ "model.language_model.layers.16.linear_attn.in_proj_b",
239
+ "model.language_model.layers.16.linear_attn.in_proj_a",
240
+ "model.language_model.layers.17.linear_attn",
241
+ "model.language_model.layers.17.linear_attn.norm",
242
+ "model.language_model.layers.17.linear_attn.in_proj_b",
243
+ "model.language_model.layers.17.linear_attn.in_proj_a",
244
+ "model.language_model.layers.18.linear_attn",
245
+ "model.language_model.layers.18.linear_attn.norm",
246
+ "model.language_model.layers.18.linear_attn.in_proj_b",
247
+ "model.language_model.layers.18.linear_attn.in_proj_a",
248
+ "model.language_model.layers.20.linear_attn",
249
+ "model.language_model.layers.20.linear_attn.norm",
250
+ "model.language_model.layers.20.linear_attn.in_proj_b",
251
+ "model.language_model.layers.20.linear_attn.in_proj_a",
252
+ "model.language_model.layers.21.linear_attn",
253
+ "model.language_model.layers.21.linear_attn.norm",
254
+ "model.language_model.layers.21.linear_attn.in_proj_b",
255
+ "model.language_model.layers.21.linear_attn.in_proj_a",
256
+ "model.language_model.layers.22.linear_attn",
257
+ "model.language_model.layers.22.linear_attn.norm",
258
+ "model.language_model.layers.22.linear_attn.in_proj_b",
259
+ "model.language_model.layers.22.linear_attn.in_proj_a",
260
+ "model.language_model.layers.24.linear_attn",
261
+ "model.language_model.layers.24.linear_attn.norm",
262
+ "model.language_model.layers.24.linear_attn.in_proj_b",
263
+ "model.language_model.layers.24.linear_attn.in_proj_a",
264
+ "model.language_model.layers.25.linear_attn",
265
+ "model.language_model.layers.25.linear_attn.norm",
266
+ "model.language_model.layers.25.linear_attn.in_proj_b",
267
+ "model.language_model.layers.25.linear_attn.in_proj_a",
268
+ "model.language_model.layers.26.linear_attn",
269
+ "model.language_model.layers.26.linear_attn.norm",
270
+ "model.language_model.layers.26.linear_attn.in_proj_b",
271
+ "model.language_model.layers.26.linear_attn.in_proj_a",
272
+ "model.language_model.layers.28.linear_attn",
273
+ "model.language_model.layers.28.linear_attn.norm",
274
+ "model.language_model.layers.28.linear_attn.in_proj_b",
275
+ "model.language_model.layers.28.linear_attn.in_proj_a",
276
+ "model.language_model.layers.29.linear_attn",
277
+ "model.language_model.layers.29.linear_attn.norm",
278
+ "model.language_model.layers.29.linear_attn.in_proj_b",
279
+ "model.language_model.layers.29.linear_attn.in_proj_a",
280
+ "model.language_model.layers.30.linear_attn",
281
+ "model.language_model.layers.30.linear_attn.norm",
282
+ "model.language_model.layers.30.linear_attn.in_proj_b",
283
+ "model.language_model.layers.30.linear_attn.in_proj_a",
284
+ "model.language_model.layers.32.linear_attn",
285
+ "model.language_model.layers.32.linear_attn.norm",
286
+ "model.language_model.layers.32.linear_attn.in_proj_b",
287
+ "model.language_model.layers.32.linear_attn.in_proj_a",
288
+ "model.language_model.layers.33.linear_attn",
289
+ "model.language_model.layers.33.linear_attn.norm",
290
+ "model.language_model.layers.33.linear_attn.in_proj_b",
291
+ "model.language_model.layers.33.linear_attn.in_proj_a",
292
+ "model.language_model.layers.34.linear_attn",
293
+ "model.language_model.layers.34.linear_attn.norm",
294
+ "model.language_model.layers.34.linear_attn.in_proj_b",
295
+ "model.language_model.layers.34.linear_attn.in_proj_a",
296
+ "model.language_model.layers.36.linear_attn",
297
+ "model.language_model.layers.36.linear_attn.norm",
298
+ "model.language_model.layers.36.linear_attn.in_proj_b",
299
+ "model.language_model.layers.36.linear_attn.in_proj_a",
300
+ "model.language_model.layers.37.linear_attn",
301
+ "model.language_model.layers.37.linear_attn.norm",
302
+ "model.language_model.layers.37.linear_attn.in_proj_b",
303
+ "model.language_model.layers.37.linear_attn.in_proj_a",
304
+ "model.language_model.layers.38.linear_attn",
305
+ "model.language_model.layers.38.linear_attn.norm",
306
+ "model.language_model.layers.38.linear_attn.in_proj_b",
307
+ "model.language_model.layers.38.linear_attn.in_proj_a",
308
+ "model.language_model.layers.40.linear_attn",
309
+ "model.language_model.layers.40.linear_attn.norm",
310
+ "model.language_model.layers.40.linear_attn.in_proj_b",
311
+ "model.language_model.layers.40.linear_attn.in_proj_a",
312
+ "model.language_model.layers.41.linear_attn",
313
+ "model.language_model.layers.41.linear_attn.norm",
314
+ "model.language_model.layers.41.linear_attn.in_proj_b",
315
+ "model.language_model.layers.41.linear_attn.in_proj_a",
316
+ "model.language_model.layers.42.linear_attn",
317
+ "model.language_model.layers.42.linear_attn.norm",
318
+ "model.language_model.layers.42.linear_attn.in_proj_b",
319
+ "model.language_model.layers.42.linear_attn.in_proj_a",
320
+ "model.language_model.layers.44.linear_attn",
321
+ "model.language_model.layers.44.linear_attn.norm",
322
+ "model.language_model.layers.44.linear_attn.in_proj_b",
323
+ "model.language_model.layers.44.linear_attn.in_proj_a",
324
+ "model.language_model.layers.45.linear_attn",
325
+ "model.language_model.layers.45.linear_attn.norm",
326
+ "model.language_model.layers.45.linear_attn.in_proj_b",
327
+ "model.language_model.layers.45.linear_attn.in_proj_a",
328
+ "model.language_model.layers.46.linear_attn",
329
+ "model.language_model.layers.46.linear_attn.norm",
330
+ "model.language_model.layers.46.linear_attn.in_proj_b",
331
+ "model.language_model.layers.46.linear_attn.in_proj_a",
332
+ "model.language_model.layers.48.linear_attn",
333
+ "model.language_model.layers.48.linear_attn.norm",
334
+ "model.language_model.layers.48.linear_attn.in_proj_b",
335
+ "model.language_model.layers.48.linear_attn.in_proj_a",
336
+ "model.language_model.layers.49.linear_attn",
337
+ "model.language_model.layers.49.linear_attn.norm",
338
+ "model.language_model.layers.49.linear_attn.in_proj_b",
339
+ "model.language_model.layers.49.linear_attn.in_proj_a",
340
+ "model.language_model.layers.50.linear_attn",
341
+ "model.language_model.layers.50.linear_attn.norm",
342
+ "model.language_model.layers.50.linear_attn.in_proj_b",
343
+ "model.language_model.layers.50.linear_attn.in_proj_a",
344
+ "model.language_model.layers.52.linear_attn",
345
+ "model.language_model.layers.52.linear_attn.norm",
346
+ "model.language_model.layers.52.linear_attn.in_proj_b",
347
+ "model.language_model.layers.52.linear_attn.in_proj_a",
348
+ "model.language_model.layers.53.linear_attn",
349
+ "model.language_model.layers.53.linear_attn.norm",
350
+ "model.language_model.layers.53.linear_attn.in_proj_b",
351
+ "model.language_model.layers.53.linear_attn.in_proj_a",
352
+ "model.language_model.layers.54.linear_attn",
353
+ "model.language_model.layers.54.linear_attn.norm",
354
+ "model.language_model.layers.54.linear_attn.in_proj_b",
355
+ "model.language_model.layers.54.linear_attn.in_proj_a",
356
+ "model.language_model.layers.56.linear_attn",
357
+ "model.language_model.layers.56.linear_attn.norm",
358
+ "model.language_model.layers.56.linear_attn.in_proj_b",
359
+ "model.language_model.layers.56.linear_attn.in_proj_a",
360
+ "model.language_model.layers.57.linear_attn",
361
+ "model.language_model.layers.57.linear_attn.norm",
362
+ "model.language_model.layers.57.linear_attn.in_proj_b",
363
+ "model.language_model.layers.57.linear_attn.in_proj_a",
364
+ "model.language_model.layers.58.linear_attn",
365
+ "model.language_model.layers.58.linear_attn.norm",
366
+ "model.language_model.layers.58.linear_attn.in_proj_b",
367
+ "model.language_model.layers.58.linear_attn.in_proj_a",
368
+ "model.language_model.layers.60.linear_attn",
369
+ "model.language_model.layers.60.linear_attn.norm",
370
+ "model.language_model.layers.60.linear_attn.in_proj_b",
371
+ "model.language_model.layers.60.linear_attn.in_proj_a",
372
+ "model.language_model.layers.61.linear_attn",
373
+ "model.language_model.layers.61.linear_attn.norm",
374
+ "model.language_model.layers.61.linear_attn.in_proj_b",
375
+ "model.language_model.layers.61.linear_attn.in_proj_a",
376
+ "model.language_model.layers.62.linear_attn",
377
+ "model.language_model.layers.62.linear_attn.norm",
378
+ "model.language_model.layers.62.linear_attn.in_proj_b",
379
+ "model.language_model.layers.62.linear_attn.in_proj_a",
380
+ "lm_head",
381
+ "re:^mtp\\."
382
+ ],
383
+ "kv_cache_scheme": {
384
+ "actorder": null,
385
+ "block_structure": null,
386
+ "dynamic": false,
387
+ "group_size": null,
388
+ "num_bits": 8,
389
+ "observer": "static_minmax",
390
+ "observer_kwargs": {},
391
+ "scale_dtype": null,
392
+ "strategy": "tensor",
393
+ "symmetric": true,
394
+ "type": "float",
395
+ "zp_dtype": null
396
+ },
397
+ "quant_method": "compressed-tensors",
398
+ "quantization_status": "compressed",
399
+ "sparsity_config": {},
400
+ "transform_config": {},
401
+ "version": "0.18.0"
402
+ },
403
+ "text_config": {
404
+ "attention_bias": false,
405
+ "attention_dropout": 0.0,
406
+ "attn_output_gate": true,
407
+ "bos_token_id": 248044,
408
+ "dtype": "bfloat16",
409
+ "eos_token_id": 248044,
410
+ "full_attention_interval": 4,
411
+ "head_dim": 256,
412
+ "hidden_act": "silu",
413
+ "hidden_size": 5120,
414
+ "initializer_range": 0.02,
415
+ "intermediate_size": 17408,
416
+ "layer_types": [
417
+ "linear_attention",
418
+ "linear_attention",
419
+ "linear_attention",
420
+ "full_attention",
421
+ "linear_attention",
422
+ "linear_attention",
423
+ "linear_attention",
424
+ "full_attention",
425
+ "linear_attention",
426
+ "linear_attention",
427
+ "linear_attention",
428
+ "full_attention",
429
+ "linear_attention",
430
+ "linear_attention",
431
+ "linear_attention",
432
+ "full_attention",
433
+ "linear_attention",
434
+ "linear_attention",
435
+ "linear_attention",
436
+ "full_attention",
437
+ "linear_attention",
438
+ "linear_attention",
439
+ "linear_attention",
440
+ "full_attention",
441
+ "linear_attention",
442
+ "linear_attention",
443
+ "linear_attention",
444
+ "full_attention",
445
+ "linear_attention",
446
+ "linear_attention",
447
+ "linear_attention",
448
+ "full_attention",
449
+ "linear_attention",
450
+ "linear_attention",
451
+ "linear_attention",
452
+ "full_attention",
453
+ "linear_attention",
454
+ "linear_attention",
455
+ "linear_attention",
456
+ "full_attention",
457
+ "linear_attention",
458
+ "linear_attention",
459
+ "linear_attention",
460
+ "full_attention",
461
+ "linear_attention",
462
+ "linear_attention",
463
+ "linear_attention",
464
+ "full_attention",
465
+ "linear_attention",
466
+ "linear_attention",
467
+ "linear_attention",
468
+ "full_attention",
469
+ "linear_attention",
470
+ "linear_attention",
471
+ "linear_attention",
472
+ "full_attention",
473
+ "linear_attention",
474
+ "linear_attention",
475
+ "linear_attention",
476
+ "full_attention",
477
+ "linear_attention",
478
+ "linear_attention",
479
+ "linear_attention",
480
+ "full_attention"
481
+ ],
482
+ "linear_conv_kernel_dim": 4,
483
+ "linear_key_head_dim": 128,
484
+ "linear_num_key_heads": 16,
485
+ "linear_num_value_heads": 48,
486
+ "linear_value_head_dim": 128,
487
+ "mamba_ssm_dtype": "float32",
488
+ "max_position_embeddings": 262144,
489
+ "model_type": "qwen3_5_text",
490
+ "mtp_num_hidden_layers": 1,
491
+ "mtp_use_dedicated_embeddings": false,
492
+ "num_attention_heads": 24,
493
+ "num_hidden_layers": 64,
494
+ "num_key_value_heads": 4,
495
+ "output_gate_type": "swish",
496
+ "pad_token_id": null,
497
+ "partial_rotary_factor": 0.25,
498
+ "rms_norm_eps": 1e-06,
499
+ "rope_parameters": {
500
+ "mrope_interleaved": true,
501
+ "mrope_section": [
502
+ 11,
503
+ 11,
504
+ 10
505
+ ],
506
+ "partial_rotary_factor": 0.25,
507
+ "rope_theta": 10000000,
508
+ "rope_type": "default"
509
+ },
510
+ "router_aux_loss_coef": 0.001,
511
+ "tie_word_embeddings": false,
512
+ "use_cache": true,
513
+ "vocab_size": 248320
514
+ },
515
+ "tie_word_embeddings": false,
516
+ "transformers_version": "5.10.1",
517
+ "video_token_id": 248057,
518
+ "vision_config": {
519
+ "deepstack_visual_indexes": [],
520
+ "depth": 27,
521
+ "dtype": "bfloat16",
522
+ "hidden_act": "gelu_pytorch_tanh",
523
+ "hidden_size": 1152,
524
+ "in_channels": 3,
525
+ "initializer_range": 0.02,
526
+ "intermediate_size": 4304,
527
+ "model_type": "qwen3_5_vision",
528
+ "num_heads": 16,
529
+ "num_position_embeddings": 2304,
530
+ "out_hidden_size": 5120,
531
+ "patch_size": 16,
532
+ "spatial_merge_size": 2,
533
+ "temporal_patch_size": 2
534
+ },
535
+ "vision_end_token_id": 248054,
536
+ "vision_start_token_id": 248053
537
+ }
crystal-mtp-provenance.json ADDED
@@ -0,0 +1,355 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema": "crystal.holo4-mtp/v1",
3
+ "inputs": {
4
+ "schema": "crystal.holo4-mtp-inputs/v1",
5
+ "sources": {
6
+ "holo": {
7
+ "repository": "Hcompany/Holo4-27B-NVFP4",
8
+ "revision": "6a02909667ba91dcd45e01be13c4d49992112510",
9
+ "files": [
10
+ {
11
+ "name": ".gitattributes",
12
+ "size": 1570,
13
+ "git_blob_sha1": "52373fe24473b1aa44333d318f578ae6bf04b49b"
14
+ },
15
+ {
16
+ "name": "LICENSE",
17
+ "size": 19347,
18
+ "git_blob_sha1": "fe463e0f7888bbf8b82e42d55f5743508ddafb7e"
19
+ },
20
+ {
21
+ "name": "LICENSE-APACHE",
22
+ "size": 11544,
23
+ "git_blob_sha1": "f938136e3adacfd92be087f6e113b5d6d97f678f"
24
+ },
25
+ {
26
+ "name": "README.md",
27
+ "size": 6765,
28
+ "git_blob_sha1": "1e7b4e538a73f23f62c2e1e890e6303b2c461529"
29
+ },
30
+ {
31
+ "name": "chat_template.jinja",
32
+ "size": 8952,
33
+ "git_blob_sha1": "c0c686f9c38d70d179fb7b5f5aa7530bc913dda3"
34
+ },
35
+ {
36
+ "name": "config.json",
37
+ "size": 22259,
38
+ "git_blob_sha1": "2409f3cd57a259d703b55d79e60c63500d9d1640"
39
+ },
40
+ {
41
+ "name": "generation_config.json",
42
+ "size": 214,
43
+ "git_blob_sha1": "8a2e2eff36818da35d959e8937fd8025def46ffa"
44
+ },
45
+ {
46
+ "name": "model-00001-of-00002.safetensors",
47
+ "size": 19983233952,
48
+ "sha256": "2cdf9e105d274204894a086d44dc805d1a4b428b9e9d8220c58b5a21972ce54f"
49
+ },
50
+ {
51
+ "name": "model-00002-of-00002.safetensors",
52
+ "size": 2916688264,
53
+ "sha256": "229ba26491580906dfdf595c9ad4a73b1fd258895fce67f3cde3f2ddb595ff03"
54
+ },
55
+ {
56
+ "name": "model.safetensors.index.json",
57
+ "size": 199734,
58
+ "git_blob_sha1": "02befdbcfea1f835e67e5c85ab4d75965af1dc56"
59
+ },
60
+ {
61
+ "name": "processor_config.json",
62
+ "size": 1191,
63
+ "git_blob_sha1": "33818c7f9e991ad735fd240209f4fa73e6c28c50"
64
+ },
65
+ {
66
+ "name": "recipe.yaml",
67
+ "size": 2262,
68
+ "git_blob_sha1": "6b6f27f2a244573d10b1cfe504c8f711ef479a39"
69
+ },
70
+ {
71
+ "name": "tokenizer.json",
72
+ "size": 19989325,
73
+ "sha256": "06b9509352d2af50381ab2247e083b80d32d5c0aba91c272ca9ff729b6a0e523"
74
+ },
75
+ {
76
+ "name": "tokenizer_config.json",
77
+ "size": 1165,
78
+ "git_blob_sha1": "1d134cd298be1e3be25db393d93a1cefe80e3214"
79
+ }
80
+ ]
81
+ },
82
+ "donor": {
83
+ "repository": "Qwen/Qwen3.8-27B",
84
+ "revision": "1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0",
85
+ "files": [
86
+ {
87
+ "name": "LICENSE",
88
+ "size": 11544,
89
+ "git_blob_sha1": "f938136e3adacfd92be087f6e113b5d6d97f678f"
90
+ },
91
+ {
92
+ "name": "README.md",
93
+ "size": 65012,
94
+ "git_blob_sha1": "bc8aa0e396cd029c21cd773cca21830d0ded28ec"
95
+ },
96
+ {
97
+ "name": "config.json",
98
+ "size": 4312,
99
+ "git_blob_sha1": "706cebd746c4b6f2b1d1f892630867acfdfd3df8"
100
+ },
101
+ {
102
+ "name": "model-00018-of-00018.safetensors",
103
+ "size": 3392197344,
104
+ "sha256": "1d3479509e21494658f9b64d317f5ea8e55c4025d28c702d6c4d0b356ce8ea06"
105
+ },
106
+ {
107
+ "name": "model.safetensors.index.json",
108
+ "size": 112216,
109
+ "git_blob_sha1": "da35e3c564457dface7d138f0b6cac284ff8958c"
110
+ }
111
+ ]
112
+ }
113
+ }
114
+ },
115
+ "verified_files": {
116
+ "holo": [
117
+ {
118
+ "name": ".gitattributes",
119
+ "size": 1570,
120
+ "sha256": "34448b82c17d60fec9b65b1f093c115ddbaadc04beb1b0140b6bfed2e012a930"
121
+ },
122
+ {
123
+ "name": "LICENSE",
124
+ "size": 19347,
125
+ "sha256": "41003d4a74749c0220e33dd415042164b5a1093ed401f36277234f772d22d3d0"
126
+ },
127
+ {
128
+ "name": "LICENSE-APACHE",
129
+ "size": 11544,
130
+ "sha256": "bbedc3fda3305820b977265f01b8619d87570a6739de3a5582c3464840f1e57a"
131
+ },
132
+ {
133
+ "name": "README.md",
134
+ "size": 6765,
135
+ "sha256": "5311535fee699a112d424787f44240549253b6d035f80872dcb8c7d5bc0fb8c5"
136
+ },
137
+ {
138
+ "name": "chat_template.jinja",
139
+ "size": 8952,
140
+ "sha256": "c3cf9e34abf4f9e36c2d72165aa9c132d3e2a725b6c2586aaa3a8af9d7a81041"
141
+ },
142
+ {
143
+ "name": "config.json",
144
+ "size": 22259,
145
+ "sha256": "f720dc90bafdd1fa80fd4670b2224182977849809cac06173e49b807d54cb0c1"
146
+ },
147
+ {
148
+ "name": "generation_config.json",
149
+ "size": 214,
150
+ "sha256": "56147698c439ec2da22f5820fcf76fa14788b6c0ebdd2b2e073c6cddfdd1ea30"
151
+ },
152
+ {
153
+ "name": "model-00001-of-00002.safetensors",
154
+ "size": 19983233952,
155
+ "sha256": "2cdf9e105d274204894a086d44dc805d1a4b428b9e9d8220c58b5a21972ce54f"
156
+ },
157
+ {
158
+ "name": "model-00002-of-00002.safetensors",
159
+ "size": 2916688264,
160
+ "sha256": "229ba26491580906dfdf595c9ad4a73b1fd258895fce67f3cde3f2ddb595ff03"
161
+ },
162
+ {
163
+ "name": "model.safetensors.index.json",
164
+ "size": 199734,
165
+ "sha256": "a7196edd8a12d82def4bab252c4a87c59f917d191252f8ea1f0760c8c3e012bc"
166
+ },
167
+ {
168
+ "name": "processor_config.json",
169
+ "size": 1191,
170
+ "sha256": "d89ef49ce9cd37fbf510158e13c1ef063d9286411c1ec9049932dbe0487143b1"
171
+ },
172
+ {
173
+ "name": "recipe.yaml",
174
+ "size": 2262,
175
+ "sha256": "f79d08cba8557097a26c6fc4713f9d214b2ba7850105d98d77f9fc9ed232d22f"
176
+ },
177
+ {
178
+ "name": "tokenizer.json",
179
+ "size": 19989325,
180
+ "sha256": "06b9509352d2af50381ab2247e083b80d32d5c0aba91c272ca9ff729b6a0e523"
181
+ },
182
+ {
183
+ "name": "tokenizer_config.json",
184
+ "size": 1165,
185
+ "sha256": "792fa3f0cb88b111e54ef3134c873531008c4df471d108da17903426e308aa7b"
186
+ }
187
+ ],
188
+ "donor": [
189
+ {
190
+ "name": "LICENSE",
191
+ "size": 11544,
192
+ "sha256": "bbedc3fda3305820b977265f01b8619d87570a6739de3a5582c3464840f1e57a"
193
+ },
194
+ {
195
+ "name": "README.md",
196
+ "size": 65012,
197
+ "sha256": "57e4bdb258ee1a7d2635c5174ebd4e56abe392505cdb5f8bbb356b0dc4293641"
198
+ },
199
+ {
200
+ "name": "config.json",
201
+ "size": 4312,
202
+ "sha256": "191e0af232104ed8b65258cf3fb2b842e288008baca7633c11b82a1ac7203aab"
203
+ },
204
+ {
205
+ "name": "model-00018-of-00018.safetensors",
206
+ "size": 3392197344,
207
+ "sha256": "1d3479509e21494658f9b64d317f5ea8e55c4025d28c702d6c4d0b356ce8ea06"
208
+ },
209
+ {
210
+ "name": "model.safetensors.index.json",
211
+ "size": 112216,
212
+ "sha256": "77042094076611b69791a610065f28b7013b8c621795fa86ddccc8bac7d1b9df"
213
+ }
214
+ ]
215
+ },
216
+ "mtp_tensors": {
217
+ "mtp.fc.weight": {
218
+ "dtype": "BF16",
219
+ "shape": [
220
+ 5120,
221
+ 10240
222
+ ],
223
+ "sha256": "4eee377b67ec2122cf214dbe6946d16261873441f1851d64409d9c7566bb20cc",
224
+ "bytes": 104857600
225
+ },
226
+ "mtp.layers.0.input_layernorm.weight": {
227
+ "dtype": "BF16",
228
+ "shape": [
229
+ 5120
230
+ ],
231
+ "sha256": "bf04147cf2ecff83cd7b4f8b8611df55955041bc7749c95f6c9eb741e2ce7ec0",
232
+ "bytes": 10240
233
+ },
234
+ "mtp.layers.0.mlp.down_proj.weight": {
235
+ "dtype": "BF16",
236
+ "shape": [
237
+ 5120,
238
+ 17408
239
+ ],
240
+ "sha256": "2fdda9751e3b7c1444632878283116550a1824dcad691aa8366793ba21faf935",
241
+ "bytes": 178257920
242
+ },
243
+ "mtp.layers.0.mlp.gate_proj.weight": {
244
+ "dtype": "BF16",
245
+ "shape": [
246
+ 17408,
247
+ 5120
248
+ ],
249
+ "sha256": "e2fa26989b3103cc595990f53c13a295f95507d0a480276e642a084e6b1f62cc",
250
+ "bytes": 178257920
251
+ },
252
+ "mtp.layers.0.mlp.up_proj.weight": {
253
+ "dtype": "BF16",
254
+ "shape": [
255
+ 17408,
256
+ 5120
257
+ ],
258
+ "sha256": "8c6a44250d5976dc707703ea09e994eaeabef309c3faac8599e2e282eb4933ba",
259
+ "bytes": 178257920
260
+ },
261
+ "mtp.layers.0.post_attention_layernorm.weight": {
262
+ "dtype": "BF16",
263
+ "shape": [
264
+ 5120
265
+ ],
266
+ "sha256": "248f158f48a4b9b1d140e6728b77b19f132ae19de5cb60dac292742f7a695bbf",
267
+ "bytes": 10240
268
+ },
269
+ "mtp.layers.0.self_attn.k_norm.weight": {
270
+ "dtype": "BF16",
271
+ "shape": [
272
+ 256
273
+ ],
274
+ "sha256": "73723b12e89816afcc8faab883f88079f0eccf6c3208b2d7d851adc2597b3c63",
275
+ "bytes": 512
276
+ },
277
+ "mtp.layers.0.self_attn.k_proj.weight": {
278
+ "dtype": "BF16",
279
+ "shape": [
280
+ 1024,
281
+ 5120
282
+ ],
283
+ "sha256": "0ab4e67930451a2f1140e865d2e3e7e23cd752a3bb25042c36e7e546361073d0",
284
+ "bytes": 10485760
285
+ },
286
+ "mtp.layers.0.self_attn.o_proj.weight": {
287
+ "dtype": "BF16",
288
+ "shape": [
289
+ 5120,
290
+ 6144
291
+ ],
292
+ "sha256": "9165a161837a5d4489a838279805b15dd5b5827f1d3ac2f3f4067a71131a5fd7",
293
+ "bytes": 62914560
294
+ },
295
+ "mtp.layers.0.self_attn.q_norm.weight": {
296
+ "dtype": "BF16",
297
+ "shape": [
298
+ 256
299
+ ],
300
+ "sha256": "76aae220c35e2dea96f0985239b719fdcab93e208add6c32ca399b1166ec444e",
301
+ "bytes": 512
302
+ },
303
+ "mtp.layers.0.self_attn.q_proj.weight": {
304
+ "dtype": "BF16",
305
+ "shape": [
306
+ 12288,
307
+ 5120
308
+ ],
309
+ "sha256": "8b27ef96b91e0b0564f82f11d9010efee6aa40a0eaa1900336bf597f767e7c98",
310
+ "bytes": 125829120
311
+ },
312
+ "mtp.layers.0.self_attn.v_proj.weight": {
313
+ "dtype": "BF16",
314
+ "shape": [
315
+ 1024,
316
+ 5120
317
+ ],
318
+ "sha256": "9606be457a48d7d9a78d72f255246f4e346845d752c1e82e41e14728255c5e13",
319
+ "bytes": 10485760
320
+ },
321
+ "mtp.norm.weight": {
322
+ "dtype": "BF16",
323
+ "shape": [
324
+ 5120
325
+ ],
326
+ "sha256": "414fb74a971b168fb9a073c3486cbc61c46e992757179e724a345bbf56896fb1",
327
+ "bytes": 10240
328
+ },
329
+ "mtp.pre_fc_norm_embedding.weight": {
330
+ "dtype": "BF16",
331
+ "shape": [
332
+ 5120
333
+ ],
334
+ "sha256": "9b03e3af7824cd765abdcfc21da960c25f3c7ee1294ce25c226f0ebc5ff39998",
335
+ "bytes": 10240
336
+ },
337
+ "mtp.pre_fc_norm_hidden.weight": {
338
+ "dtype": "BF16",
339
+ "shape": [
340
+ 5120
341
+ ],
342
+ "sha256": "f2e1aa414ce84d89d016f2260923f7d47eb9827be043d1823dc6e783e4205efe",
343
+ "bytes": 10240
344
+ }
345
+ },
346
+ "target_layers": 64,
347
+ "mtp_layers": 1,
348
+ "target_weight_changes": [],
349
+ "quantization_ignore_added": "re:^mtp\\.",
350
+ "composite_sha256": {
351
+ "config.json": "8fe4a28d2437e47e12246b1a5e4415f848e4dc89fb86ccfcd1cd32a8ba09f52f",
352
+ "model.safetensors.index.json": "4a701c716a3b5616eafba7402e70907ef8f7e0266d929d7bd14417d58cb0dc98",
353
+ "model-mtp.safetensors": "90fa0e3eed5a647c035c6df9ecabc416c0f8d573ff84ac12485b085f00a7cdf2"
354
+ }
355
+ }
generation_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "transformers_version": "5.10.1"
13
+ }
model-00001-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2cdf9e105d274204894a086d44dc805d1a4b428b9e9d8220c58b5a21972ce54f
3
+ size 19983233952
model-00002-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:229ba26491580906dfdf595c9ad4a73b1fd258895fce67f3cde3f2ddb595ff03
3
+ size 2916688264
model-mtp.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:90fa0e3eed5a647c035c6df9ecabc416c0f8d573ff84ac12485b085f00a7cdf2
3
+ size 849400424
model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
processor_config.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "merge_size": 2,
19
+ "patch_size": 16,
20
+ "resample": 3,
21
+ "rescale_factor": 0.00392156862745098,
22
+ "size": {
23
+ "longest_edge": 16777216,
24
+ "shortest_edge": 65536
25
+ },
26
+ "temporal_patch_size": 2
27
+ },
28
+ "processor_class": "Qwen3VLProcessor",
29
+ "video_processor": {
30
+ "do_convert_rgb": true,
31
+ "do_normalize": true,
32
+ "do_rescale": true,
33
+ "do_resize": true,
34
+ "do_sample_frames": true,
35
+ "fps": 2,
36
+ "image_mean": [
37
+ 0.5,
38
+ 0.5,
39
+ 0.5
40
+ ],
41
+ "image_std": [
42
+ 0.5,
43
+ 0.5,
44
+ 0.5
45
+ ],
46
+ "max_frames": 768,
47
+ "merge_size": 2,
48
+ "min_frames": 4,
49
+ "patch_size": 16,
50
+ "resample": 3,
51
+ "rescale_factor": 0.00392156862745098,
52
+ "return_metadata": false,
53
+ "size": {
54
+ "longest_edge": 25165824,
55
+ "shortest_edge": 4096
56
+ },
57
+ "temporal_patch_size": 2,
58
+ "video_processor_type": "Qwen3VLVideoProcessor"
59
+ }
60
+ }
recipe.yaml ADDED
@@ -0,0 +1,71 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ default_stage:
2
+ default_modifiers:
3
+ QuantizationModifier:
4
+ config_groups:
5
+ experts_nvfp4a16:
6
+ targets: ['re:.*mlp\.experts\.\d+\.(gate|up|down)_proj$', 're:.*mlp\.shared_expert\.(gate|up|down)_proj$',
7
+ 're:.*mlp\.(gate|up|down)_proj$']
8
+ weights:
9
+ num_bits: 4
10
+ type: float
11
+ symmetric: true
12
+ group_size: 16
13
+ strategy: tensor_group
14
+ block_structure: null
15
+ dynamic: false
16
+ actorder: null
17
+ scale_dtype: torch.float8_e4m3fn
18
+ zp_dtype: null
19
+ observer: memoryless_minmax
20
+ observer_kwargs: {}
21
+ input_activations: null
22
+ output_activations: null
23
+ format: null
24
+ attention_fp8_static:
25
+ targets: ['re:.*self_attn\.(q|k|v|o)_proj$', 're:.*linear_attn\.(in_proj_qkv|in_proj_z|out_proj)$']
26
+ weights:
27
+ num_bits: 8
28
+ type: float
29
+ symmetric: true
30
+ group_size: null
31
+ strategy: tensor
32
+ block_structure: null
33
+ dynamic: false
34
+ actorder: null
35
+ scale_dtype: null
36
+ zp_dtype: null
37
+ observer: memoryless_minmax
38
+ observer_kwargs: {}
39
+ input_activations:
40
+ num_bits: 8
41
+ type: float
42
+ symmetric: true
43
+ group_size: null
44
+ strategy: tensor
45
+ block_structure: null
46
+ dynamic: false
47
+ actorder: null
48
+ scale_dtype: null
49
+ zp_dtype: null
50
+ observer: static_minmax
51
+ observer_kwargs: {}
52
+ output_activations: null
53
+ format: null
54
+ targets: [Linear]
55
+ ignore: [lm_head, 're:.*visual\..*', 're:.*mlp\.gate$', 're:.*mlp\.shared_expert_gate$',
56
+ 're:.*embed_tokens$', 're:.*mtp\..*']
57
+ kv_cache_scheme:
58
+ num_bits: 8
59
+ type: float
60
+ symmetric: true
61
+ group_size: null
62
+ strategy: tensor
63
+ block_structure: null
64
+ dynamic: false
65
+ actorder: null
66
+ scale_dtype: null
67
+ zp_dtype: null
68
+ observer: static_minmax
69
+ observer_kwargs: {}
70
+ bypass_divisibility_checks: false
71
+ requires_calibration_data: true
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:06b9509352d2af50381ab2247e083b80d32d5c0aba91c272ca9ff729b6a0e523
3
+ size 19989325
tokenizer_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "local_files_only": false,
14
+ "model_max_length": 262144,
15
+ "model_specific_special_tokens": {
16
+ "audio_bos_token": "<|audio_start|>",
17
+ "audio_eos_token": "<|audio_end|>",
18
+ "audio_token": "<|audio_pad|>",
19
+ "image_token": "<|image_pad|>",
20
+ "video_token": "<|video_pad|>",
21
+ "vision_bos_token": "<|vision_start|>",
22
+ "vision_eos_token": "<|vision_end|>"
23
+ },
24
+ "pad_token": "<|endoftext|>",
25
+ "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+",
26
+ "processor_class": "Qwen3VLProcessor",
27
+ "split_special_tokens": false,
28
+ "tokenizer_class": "Qwen2Tokenizer",
29
+ "unk_token": null,
30
+ "video_token": "<|video_pad|>",
31
+ "vision_bos_token": "<|vision_start|>",
32
+ "vision_eos_token": "<|vision_end|>"
33
+ }