Commit ·
c25cc08
unverified ·
0
Parent(s):
Initial commit
Browse files- .gitattributes +46 -0
- LICENSE +202 -0
- README.md +967 -0
- TYPES.md +36 -0
- chat_template.jinja +139 -0
- config.json +112 -0
- extraction_performance_no_thinking.png +3 -0
- extraction_performance_thinking.png +3 -0
- generation_config.json +6 -0
- header.svg +0 -0
- logo.png +3 -0
- logo_nuextract.svg +95 -0
- markdown.png +3 -0
- markdown_base.png +3 -0
- markdown_reasoning.png +3 -0
- md2st.svg +139 -0
- model.safetensors +3 -0
- model.safetensors.index.json +745 -0
- model_mtp.safetensors +3 -0
- ocr_benchmark_specialized.png +3 -0
- ocr_preferences.svg +154 -0
- processor_config.json +60 -0
- st.svg +347 -0
- tokenizer.json +3 -0
- tokenizer_config.json +33 -0
.gitattributes
ADDED
|
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
*.7z filter=lfs diff=lfs merge=lfs -text
|
| 2 |
+
*.arrow filter=lfs diff=lfs merge=lfs -text
|
| 3 |
+
*.bin filter=lfs diff=lfs merge=lfs -text
|
| 4 |
+
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
| 5 |
+
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
| 6 |
+
*.ftz filter=lfs diff=lfs merge=lfs -text
|
| 7 |
+
*.gz filter=lfs diff=lfs merge=lfs -text
|
| 8 |
+
*.h5 filter=lfs diff=lfs merge=lfs -text
|
| 9 |
+
*.joblib filter=lfs diff=lfs merge=lfs -text
|
| 10 |
+
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
| 11 |
+
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
| 12 |
+
*.model filter=lfs diff=lfs merge=lfs -text
|
| 13 |
+
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
| 14 |
+
*.npy filter=lfs diff=lfs merge=lfs -text
|
| 15 |
+
*.npz filter=lfs diff=lfs merge=lfs -text
|
| 16 |
+
*.onnx filter=lfs diff=lfs merge=lfs -text
|
| 17 |
+
*.ot filter=lfs diff=lfs merge=lfs -text
|
| 18 |
+
*.parquet filter=lfs diff=lfs merge=lfs -text
|
| 19 |
+
*.pb filter=lfs diff=lfs merge=lfs -text
|
| 20 |
+
*.pickle filter=lfs diff=lfs merge=lfs -text
|
| 21 |
+
*.pkl filter=lfs diff=lfs merge=lfs -text
|
| 22 |
+
*.pt filter=lfs diff=lfs merge=lfs -text
|
| 23 |
+
*.pth filter=lfs diff=lfs merge=lfs -text
|
| 24 |
+
*.rar filter=lfs diff=lfs merge=lfs -text
|
| 25 |
+
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
| 26 |
+
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
| 27 |
+
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
| 28 |
+
*.tar filter=lfs diff=lfs merge=lfs -text
|
| 29 |
+
*.tflite filter=lfs diff=lfs merge=lfs -text
|
| 30 |
+
*.tgz filter=lfs diff=lfs merge=lfs -text
|
| 31 |
+
*.wasm filter=lfs diff=lfs merge=lfs -text
|
| 32 |
+
*.xz filter=lfs diff=lfs merge=lfs -text
|
| 33 |
+
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
+
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
+
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
markdown_base.png filter=lfs diff=lfs merge=lfs -text
|
| 38 |
+
markdown_reasoning.png filter=lfs diff=lfs merge=lfs -text
|
| 39 |
+
structured.png filter=lfs diff=lfs merge=lfs -text
|
| 40 |
+
markdown.png filter=lfs diff=lfs merge=lfs -text
|
| 41 |
+
extraction_performance_no_thinking.png filter=lfs diff=lfs merge=lfs -text
|
| 42 |
+
extraction_performance_thinking.png filter=lfs diff=lfs merge=lfs -text
|
| 43 |
+
logo.png filter=lfs diff=lfs merge=lfs -text
|
| 44 |
+
structured-nothinking.png filter=lfs diff=lfs merge=lfs -text
|
| 45 |
+
structured-thinking.png filter=lfs diff=lfs merge=lfs -text
|
| 46 |
+
ocr_benchmark_specialized.png filter=lfs diff=lfs merge=lfs -text
|
LICENSE
ADDED
|
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
Apache License
|
| 3 |
+
Version 2.0, January 2004
|
| 4 |
+
http://www.apache.org/licenses/
|
| 5 |
+
|
| 6 |
+
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
|
| 7 |
+
|
| 8 |
+
1. Definitions.
|
| 9 |
+
|
| 10 |
+
"License" shall mean the terms and conditions for use, reproduction,
|
| 11 |
+
and distribution as defined by Sections 1 through 9 of this document.
|
| 12 |
+
|
| 13 |
+
"Licensor" shall mean the copyright owner or entity authorized by
|
| 14 |
+
the copyright owner that is granting the License.
|
| 15 |
+
|
| 16 |
+
"Legal Entity" shall mean the union of the acting entity and all
|
| 17 |
+
other entities that control, are controlled by, or are under common
|
| 18 |
+
control with that entity. For the purposes of this definition,
|
| 19 |
+
"control" means (i) the power, direct or indirect, to cause the
|
| 20 |
+
direction or management of such entity, whether by contract or
|
| 21 |
+
otherwise, or (ii) ownership of fifty percent (50%) or more of the
|
| 22 |
+
outstanding shares, or (iii) beneficial ownership of such entity.
|
| 23 |
+
|
| 24 |
+
"You" (or "Your") shall mean an individual or Legal Entity
|
| 25 |
+
exercising permissions granted by this License.
|
| 26 |
+
|
| 27 |
+
"Source" form shall mean the preferred form for making modifications,
|
| 28 |
+
including but not limited to software source code, documentation
|
| 29 |
+
source, and configuration files.
|
| 30 |
+
|
| 31 |
+
"Object" form shall mean any form resulting from mechanical
|
| 32 |
+
transformation or translation of a Source form, including but
|
| 33 |
+
not limited to compiled object code, generated documentation,
|
| 34 |
+
and conversions to other media types.
|
| 35 |
+
|
| 36 |
+
"Work" shall mean the work of authorship, whether in Source or
|
| 37 |
+
Object form, made available under the License, as indicated by a
|
| 38 |
+
copyright notice that is included in or attached to the work
|
| 39 |
+
(an example is provided in the Appendix below).
|
| 40 |
+
|
| 41 |
+
"Derivative Works" shall mean any work, whether in Source or Object
|
| 42 |
+
form, that is based on (or derived from) the Work and for which the
|
| 43 |
+
editorial revisions, annotations, elaborations, or other modifications
|
| 44 |
+
represent, as a whole, an original work of authorship. For the purposes
|
| 45 |
+
of this License, Derivative Works shall not include works that remain
|
| 46 |
+
separable from, or merely link (or bind by name) to the interfaces of,
|
| 47 |
+
the Work and Derivative Works thereof.
|
| 48 |
+
|
| 49 |
+
"Contribution" shall mean any work of authorship, including
|
| 50 |
+
the original version of the Work and any modifications or additions
|
| 51 |
+
to that Work or Derivative Works thereof, that is intentionally
|
| 52 |
+
submitted to Licensor for inclusion in the Work by the copyright owner
|
| 53 |
+
or by an individual or Legal Entity authorized to submit on behalf of
|
| 54 |
+
the copyright owner. For the purposes of this definition, "submitted"
|
| 55 |
+
means any form of electronic, verbal, or written communication sent
|
| 56 |
+
to the Licensor or its representatives, including but not limited to
|
| 57 |
+
communication on electronic mailing lists, source code control systems,
|
| 58 |
+
and issue tracking systems that are managed by, or on behalf of, the
|
| 59 |
+
Licensor for the purpose of discussing and improving the Work, but
|
| 60 |
+
excluding communication that is conspicuously marked or otherwise
|
| 61 |
+
designated in writing by the copyright owner as "Not a Contribution."
|
| 62 |
+
|
| 63 |
+
"Contributor" shall mean Licensor and any individual or Legal Entity
|
| 64 |
+
on behalf of whom a Contribution has been received by Licensor and
|
| 65 |
+
subsequently incorporated within the Work.
|
| 66 |
+
|
| 67 |
+
2. Grant of Copyright License. Subject to the terms and conditions of
|
| 68 |
+
this License, each Contributor hereby grants to You a perpetual,
|
| 69 |
+
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
| 70 |
+
copyright license to reproduce, prepare Derivative Works of,
|
| 71 |
+
publicly display, publicly perform, sublicense, and distribute the
|
| 72 |
+
Work and such Derivative Works in Source or Object form.
|
| 73 |
+
|
| 74 |
+
3. Grant of Patent License. Subject to the terms and conditions of
|
| 75 |
+
this License, each Contributor hereby grants to You a perpetual,
|
| 76 |
+
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
| 77 |
+
(except as stated in this section) patent license to make, have made,
|
| 78 |
+
use, offer to sell, sell, import, and otherwise transfer the Work,
|
| 79 |
+
where such license applies only to those patent claims licensable
|
| 80 |
+
by such Contributor that are necessarily infringed by their
|
| 81 |
+
Contribution(s) alone or by combination of their Contribution(s)
|
| 82 |
+
with the Work to which such Contribution(s) was submitted. If You
|
| 83 |
+
institute patent litigation against any entity (including a
|
| 84 |
+
cross-claim or counterclaim in a lawsuit) alleging that the Work
|
| 85 |
+
or a Contribution incorporated within the Work constitutes direct
|
| 86 |
+
or contributory patent infringement, then any patent licenses
|
| 87 |
+
granted to You under this License for that Work shall terminate
|
| 88 |
+
as of the date such litigation is filed.
|
| 89 |
+
|
| 90 |
+
4. Redistribution. You may reproduce and distribute copies of the
|
| 91 |
+
Work or Derivative Works thereof in any medium, with or without
|
| 92 |
+
modifications, and in Source or Object form, provided that You
|
| 93 |
+
meet the following conditions:
|
| 94 |
+
|
| 95 |
+
(a) You must give any other recipients of the Work or
|
| 96 |
+
Derivative Works a copy of this License; and
|
| 97 |
+
|
| 98 |
+
(b) You must cause any modified files to carry prominent notices
|
| 99 |
+
stating that You changed the files; and
|
| 100 |
+
|
| 101 |
+
(c) You must retain, in the Source form of any Derivative Works
|
| 102 |
+
that You distribute, all copyright, patent, trademark, and
|
| 103 |
+
attribution notices from the Source form of the Work,
|
| 104 |
+
excluding those notices that do not pertain to any part of
|
| 105 |
+
the Derivative Works; and
|
| 106 |
+
|
| 107 |
+
(d) If the Work includes a "NOTICE" text file as part of its
|
| 108 |
+
distribution, then any Derivative Works that You distribute must
|
| 109 |
+
include a readable copy of the attribution notices contained
|
| 110 |
+
within such NOTICE file, excluding those notices that do not
|
| 111 |
+
pertain to any part of the Derivative Works, in at least one
|
| 112 |
+
of the following places: within a NOTICE text file distributed
|
| 113 |
+
as part of the Derivative Works; within the Source form or
|
| 114 |
+
documentation, if provided along with the Derivative Works; or,
|
| 115 |
+
within a display generated by the Derivative Works, if and
|
| 116 |
+
wherever such third-party notices normally appear. The contents
|
| 117 |
+
of the NOTICE file are for informational purposes only and
|
| 118 |
+
do not modify the License. You may add Your own attribution
|
| 119 |
+
notices within Derivative Works that You distribute, alongside
|
| 120 |
+
or as an addendum to the NOTICE text from the Work, provided
|
| 121 |
+
that such additional attribution notices cannot be construed
|
| 122 |
+
as modifying the License.
|
| 123 |
+
|
| 124 |
+
You may add Your own copyright statement to Your modifications and
|
| 125 |
+
may provide additional or different license terms and conditions
|
| 126 |
+
for use, reproduction, or distribution of Your modifications, or
|
| 127 |
+
for any such Derivative Works as a whole, provided Your use,
|
| 128 |
+
reproduction, and distribution of the Work otherwise complies with
|
| 129 |
+
the conditions stated in this License.
|
| 130 |
+
|
| 131 |
+
5. Submission of Contributions. Unless You explicitly state otherwise,
|
| 132 |
+
any Contribution intentionally submitted for inclusion in the Work
|
| 133 |
+
by You to the Licensor shall be under the terms and conditions of
|
| 134 |
+
this License, without any additional terms or conditions.
|
| 135 |
+
Notwithstanding the above, nothing herein shall supersede or modify
|
| 136 |
+
the terms of any separate license agreement you may have executed
|
| 137 |
+
with Licensor regarding such Contributions.
|
| 138 |
+
|
| 139 |
+
6. Trademarks. This License does not grant permission to use the trade
|
| 140 |
+
names, trademarks, service marks, or product names of the Licensor,
|
| 141 |
+
except as required for reasonable and customary use in describing the
|
| 142 |
+
origin of the Work and reproducing the content of the NOTICE file.
|
| 143 |
+
|
| 144 |
+
7. Disclaimer of Warranty. Unless required by applicable law or
|
| 145 |
+
agreed to in writing, Licensor provides the Work (and each
|
| 146 |
+
Contributor provides its Contributions) on an "AS IS" BASIS,
|
| 147 |
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
|
| 148 |
+
implied, including, without limitation, any warranties or conditions
|
| 149 |
+
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
|
| 150 |
+
PARTICULAR PURPOSE. You are solely responsible for determining the
|
| 151 |
+
appropriateness of using or redistributing the Work and assume any
|
| 152 |
+
risks associated with Your exercise of permissions under this License.
|
| 153 |
+
|
| 154 |
+
8. Limitation of Liability. In no event and under no legal theory,
|
| 155 |
+
whether in tort (including negligence), contract, or otherwise,
|
| 156 |
+
unless required by applicable law (such as deliberate and grossly
|
| 157 |
+
negligent acts) or agreed to in writing, shall any Contributor be
|
| 158 |
+
liable to You for damages, including any direct, indirect, special,
|
| 159 |
+
incidental, or consequential damages of any character arising as a
|
| 160 |
+
result of this License or out of the use or inability to use the
|
| 161 |
+
Work (including but not limited to damages for loss of goodwill,
|
| 162 |
+
work stoppage, computer failure or malfunction, or any and all
|
| 163 |
+
other commercial damages or losses), even if such Contributor
|
| 164 |
+
has been advised of the possibility of such damages.
|
| 165 |
+
|
| 166 |
+
9. Accepting Warranty or Additional Liability. While redistributing
|
| 167 |
+
the Work or Derivative Works thereof, You may choose to offer,
|
| 168 |
+
and charge a fee for, acceptance of support, warranty, indemnity,
|
| 169 |
+
or other liability obligations and/or rights consistent with this
|
| 170 |
+
License. However, in accepting such obligations, You may act only
|
| 171 |
+
on Your own behalf and on Your sole responsibility, not on behalf
|
| 172 |
+
of any other Contributor, and only if You agree to indemnify,
|
| 173 |
+
defend, and hold each Contributor harmless for any liability
|
| 174 |
+
incurred by, or claims asserted against, such Contributor by reason
|
| 175 |
+
of your accepting any such warranty or additional liability.
|
| 176 |
+
|
| 177 |
+
END OF TERMS AND CONDITIONS
|
| 178 |
+
|
| 179 |
+
APPENDIX: How to apply the Apache License to your work.
|
| 180 |
+
|
| 181 |
+
To apply the Apache License to your work, attach the following
|
| 182 |
+
boilerplate notice, with the fields enclosed by brackets "[]"
|
| 183 |
+
replaced with your own identifying information. (Don't include
|
| 184 |
+
the brackets!) The text should be enclosed in the appropriate
|
| 185 |
+
comment syntax for the file format. We also recommend that a
|
| 186 |
+
file or class name and description of purpose be included on the
|
| 187 |
+
same "printed page" as the copyright notice for easier
|
| 188 |
+
identification within third-party archives.
|
| 189 |
+
|
| 190 |
+
Copyright 2026 Alibaba Cloud
|
| 191 |
+
|
| 192 |
+
Licensed under the Apache License, Version 2.0 (the "License");
|
| 193 |
+
you may not use this file except in compliance with the License.
|
| 194 |
+
You may obtain a copy of the License at
|
| 195 |
+
|
| 196 |
+
http://www.apache.org/licenses/LICENSE-2.0
|
| 197 |
+
|
| 198 |
+
Unless required by applicable law or agreed to in writing, software
|
| 199 |
+
distributed under the License is distributed on an "AS IS" BASIS,
|
| 200 |
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
| 201 |
+
See the License for the specific language governing permissions and
|
| 202 |
+
limitations under the License.
|
README.md
ADDED
|
@@ -0,0 +1,967 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
license_link: https://huggingface.co/numind/NuExtract3/blob/main/LICENSE
|
| 4 |
+
library_name: transformers
|
| 5 |
+
pipeline_tag: image-text-to-text
|
| 6 |
+
tags:
|
| 7 |
+
- structured-extraction
|
| 8 |
+
- OCR
|
| 9 |
+
- vision-language
|
| 10 |
+
- VLM
|
| 11 |
+
- document-to-markdown
|
| 12 |
+
- markdown
|
| 13 |
+
- extraction
|
| 14 |
+
- RAG
|
| 15 |
+
- reasoning
|
| 16 |
+
- qwen
|
| 17 |
+
base_model:
|
| 18 |
+
- Qwen/Qwen3.5-4B
|
| 19 |
+
model_name: NuExtract3
|
| 20 |
+
---
|
| 21 |
+
|
| 22 |
+
<p align="center">
|
| 23 |
+
<a href="https://nuextract.ai/">
|
| 24 |
+
<img src="header.svg" width="900px"/>
|
| 25 |
+
</a>
|
| 26 |
+
</p>
|
| 27 |
+
|
| 28 |
+
|
| 29 |
+
<p align="center">
|
| 30 |
+
🖥️ <a href="https://nuextract.ai/">API / Platform</a> |
|
| 31 |
+
📑 <a href="https://numind.ai/blog">Blog</a> |
|
| 32 |
+
🗣️ <a href="https://discord.gg/3tsEtJNCDe">Discord</a> |
|
| 33 |
+
🛠️ <a href="https://github.com/numindai/nuextract">GitHub</a>
|
| 34 |
+
</p>
|
| 35 |
+
|
| 36 |
+
**NuExtract3** is a unified **4B** vision-language reasoning model for document understanding.
|
| 37 |
+
|
| 38 |
+
It combines strong **structured information extraction** with high-quality **image-to-Markdown** conversion, making it suitable for extraction pipelines, OCR, and RAG preprocessing for all types of documents such as scans, receipts, forms, invoices, contracts or tables.
|
| 39 |
+
|
| 40 |
+
Try it out in [the 🤗 space!](https://huggingface.co/spaces/numind/NuExtract-3-4B)
|
| 41 |
+
|
| 42 |
+
## Overview
|
| 43 |
+
|
| 44 |
+
- **Structured extraction**: input (text/images) + JSON template + instructions --> JSON output
|
| 45 |
+
- **Markdown conversion**: input (text/images) --> Markdown
|
| 46 |
+
- **Multimodal inputs**: text, images, or text + images.
|
| 47 |
+
- **Multilingual** documents.
|
| 48 |
+
- **Reasoning** and non-reasoning inference modes.
|
| 49 |
+
- **Template generation** for structured extraction from natural language or input document.
|
| 50 |
+
|
| 51 |
+
# Benchmark results
|
| 52 |
+
|
| 53 |
+
## Structured Extraction
|
| 54 |
+
|
| 55 |
+
We benchmarked NuExtract on NuMind's internal structured benchmark, measuring model's performances on ~600 documents of diverse types including invoices, movie posters or floor plans. These documents and their ground-truth cover diverse use-cases testing model visual understanding, OCR, reasoning and ability to handle long input and output contexts.
|
| 56 |
+
We plan to open-source this benchmark in the coming weeks, along with a extensive leaderboard including most popular open-weight and closed-sourced APIs and a Python library allowing to easily measure model performances on structured extraction.
|
| 57 |
+
|
| 58 |
+
<img src="st.svg" width="1000"/>
|
| 59 |
+
|
| 60 |
+
To measure a pair of predicted and ground-truth JSONs, we represent both as trees which we align based on node names, compute metric scores for aligned leaves and report the average of these scores. `string` and `verbatim-string` leaves are evaluated with indel distance (i.e. Levenshtein without replacement), while all others are evaluated with exact-match.
|
| 61 |
+
Models were evaluated using vllm, with a temperature of 0.25 and a maximum of 65000 output token (for both thinking and answer), which largely exceeds 22000 which is the number of tokens of the largest ground truth output.
|
| 62 |
+
|
| 63 |
+
<figure>
|
| 64 |
+
|
| 65 |
+
|Model name |Average score|Num. failed⁽¹⁾|Avg. num tokens thinking|Avg. num tokens answer|
|
| 66 |
+
|--------------------|-------------|-----------|------------------------|----------------------|
|
| 67 |
+
|NuExtract3.4_4B-RL |**0.651 ± 0.019**|27 |2036 |1856 |
|
| 68 |
+
|gemma-4-E4B-it |0.538 ± 0.023|31 |3005 |1287 |
|
| 69 |
+
|Qwen3.5-9B |0.479 ± 0.030|170 |22409 |1257 |
|
| 70 |
+
|Qwen3.5-4B |0.417 ± 0.031|229 |27177 |1201 |
|
| 71 |
+
|GLM-4.6V-Flash |0.435 ± 0.026|153 |2989 |1357 |
|
| 72 |
+
|Nemotron-3-Nano-Omni|0.387 ± 0.028|204 |25827 |522 |
|
| 73 |
+
|Ministral-3-3B |0.240 ± 0.022|344 |27586 |362 |
|
| 74 |
+
|
| 75 |
+
<figcaption>
|
| 76 |
+
<small>
|
| 77 |
+
(1) number of model outputs that were not JSON deserializable, either directly or by removing leading and trailing backticks.<br>
|
| 78 |
+
95% confidence intervals computed using a nonparametric bootstrap over scores distributions.
|
| 79 |
+
</small>
|
| 80 |
+
</figcaption>
|
| 81 |
+
</figure>
|
| 82 |
+
|
| 83 |
+
The benchmark include samples containing multiple images resulting in large input context, and some with ground-truth containing large numbers of items to extract resulting in large outputs. We found that the reasoning of small models significantly negatively impact their performances. The reason is that many models ended up falling in repetition loops, hitting the output tokens limit and resulting in failed requests.
|
| 84 |
+
|
| 85 |
+
## Document to Markdown
|
| 86 |
+
|
| 87 |
+
NuExtract can also convert document images into clean Markdown. Output will be Markdown for text (headers etc), HTML for tables, LaTeX for math and ```<figure data-type="image" data-id="img_n"><img src="/NM-dev/model_card-A/resolve/main/img_n.png" alt="Detail description of the images"/> ```
|
| 88 |
+
|
| 89 |
+
Modern, format-agnostic benchmarks for complex document understanding are limited, so we explored a new evaluation approach.
|
| 90 |
+
We selected 100 documents with challenging layouts and tables, asked each model to convert them into a structured representation, then used Gemini 3 Flash to compare model outputs against the source document and choose the most accurate result.
|
| 91 |
+
The rankings aligned with human votes, suggesting this is a promising method for evaluating document-to-Markdown capabilities. More details will be shared in an upcoming technical report.
|
| 92 |
+
Here are some results:
|
| 93 |
+
|
| 94 |
+
<img src="ocr_preferences.svg" width="1000"/>
|
| 95 |
+
|
| 96 |
+
### Using "Markdown-to-structured"
|
| 97 |
+
|
| 98 |
+
To add other evaluate references, we used our structured extraction benchmark to evaluate models in a two-step fashion: convert the benchmark inputs to Markdown, then use Qwen3.6 27B to perform the structured extraction task on them. Intuitively, it allows to evaluate how models achieve to keep the input document content and layout: good models will allow the "structured extractor" model to perform better scores.
|
| 99 |
+
|
| 100 |
+
<img src="md2st.svg" width="1000"/>
|
| 101 |
+
|
| 102 |
+
#### No thinking
|
| 103 |
+
|
| 104 |
+
| Model | Average score | Avg. num tokens answer |
|
| 105 |
+
| ---------------------| ------------: | ---------------------: |
|
| 106 |
+
| *Generalist models* | | |
|
| 107 |
+
| NuExtract3 |**0.683 ± 0.021**|1821 |
|
| 108 |
+
| Qwen3.5-4B |0.604 ± 0.025 |1797 |
|
| 109 |
+
| gemma-4-E4B |0.502 ± 0.027 |613 |
|
| 110 |
+
| GLM-4.6V-Flash |0.579 ± 0.029 |928 |
|
| 111 |
+
| Nemotron-3-Nano-Omni |0.640 ± 0.024 |1382 |
|
| 112 |
+
| granite-vision-4.1-4b|0.468 ± 0.026 |750 |
|
| 113 |
+
| Ministral-3-3B |0.521 ± 0.025 |3248 |
|
| 114 |
+
| *Specialized OCR/Markdown models* | | |
|
| 115 |
+
| GLM-OCR |0.631 ± 0.024 |1247 |
|
| 116 |
+
| LightOnOCR-2-1B |0.633 ± 0.024 |1073 |
|
| 117 |
+
| olmOCR-2-7B-1025 |0.587 ± 0.024 |732 |
|
| 118 |
+
| chandra-ocr-2 |0.665 ± 0.021 |2012 |
|
| 119 |
+
| PaddleOCR-VL-1.5 |0.433 ± 0.025 |747 |
|
| 120 |
+
|
| 121 |
+
#### Thinking
|
| 122 |
+
| Model | Average score | Avg. num tokens thinking | Avg. num tokens answer |
|
| 123 |
+
| -------------------- | ----------------: | -----------------------: | ---------------------: |
|
| 124 |
+
| NuExtract3 | **0.701 ± 0.019** | 338 | 1981 |
|
| 125 |
+
| Qwen3.5-4B | 0.662 ± 0.022 | 6552 | 1547 |
|
| 126 |
+
| gemma-4-E4B-it | 0.550 ± 0.023 | 790 | 672 |
|
| 127 |
+
| GLM-4.6V-Flash | 0.638 ± 0.023 | 1973 | 886 |
|
| 128 |
+
| Nemotron-3-Nano-Omni | 0.626 ± 0.025 | 11725 | 1040 |
|
| 129 |
+
| Ministral-3-3B | 0.542 ± 0.026 | 7593 | 773|
|
| 130 |
+
|
| 131 |
+
|
| 132 |
+
|
| 133 |
+
# Using NuExtract
|
| 134 |
+
|
| 135 |
+
## Structured extraction
|
| 136 |
+
|
| 137 |
+
Structured extraction takes as inputs:
|
| 138 |
+
|
| 139 |
+
1. An input document, which can be text, image, or both;
|
| 140 |
+
2. A JSON template describing the information to extract;
|
| 141 |
+
3. (Optional) Instructions, allowing to specify expected output formats or values;
|
| 142 |
+
4. (Optional) In-Context Learning (ICL) examples.
|
| 143 |
+
|
| 144 |
+
### Input JSON template
|
| 145 |
+
|
| 146 |
+
NuExtract uses a input JSON template whose structure is identical to the output JSON. Its leaf values are specify the **types** of the output JSON leaves. For examples:
|
| 147 |
+
|
| 148 |
+
```json
|
| 149 |
+
{
|
| 150 |
+
"invoice_number": "verbatim-string",
|
| 151 |
+
"invoice_date": "date",
|
| 152 |
+
"total_amount": "number",
|
| 153 |
+
"currency": "currency",
|
| 154 |
+
"line_items": [
|
| 155 |
+
{
|
| 156 |
+
"description": "verbatim-string",
|
| 157 |
+
"item_type": ["electronics", "clothing", "vehicle", "furniture", "other"],
|
| 158 |
+
"quantity": "integer",
|
| 159 |
+
"unit_price": "number",
|
| 160 |
+
"total": "number"
|
| 161 |
+
}
|
| 162 |
+
]
|
| 163 |
+
}
|
| 164 |
+
```
|
| 165 |
+
|
| 166 |
+
Supported template types include:
|
| 167 |
+
|
| 168 |
+
- `verbatim-string`: extract text exactly as it appears in the document;
|
| 169 |
+
- `string`: generic string field, allowing abstraction or light paraphrasing;
|
| 170 |
+
- `integer`: whole number;
|
| 171 |
+
- `number`: integer or decimal number;
|
| 172 |
+
- `date-time`: ISO-8601 date, time or date-time;
|
| 173 |
+
- Other specific types such as `data`, `time`, `country`, `currency`, `email` and so on.
|
| 174 |
+
[**For more details, read the complete types specifications and examples**](TYPES.md)
|
| 175 |
+
|
| 176 |
+
Template constructors:
|
| 177 |
+
|
| 178 |
+
- Arrays, for example `["string"]`;
|
| 179 |
+
- Enums, for example `["yes", "no", "maybe"]`;
|
| 180 |
+
- Multi-enums (multiple possible values), for example `[["A", "B", "C"]]`.
|
| 181 |
+
|
| 182 |
+
If the model does not find relevant information for a field, it returns `null` or `[]`.
|
| 183 |
+
|
| 184 |
+
### Converting JSON schema / Pydantic models to NuExtract template
|
| 185 |
+
|
| 186 |
+
Our Python SDK (`pip install numind`) offers a method to convert JSON schemas to NuExtract templates:
|
| 187 |
+
|
| 188 |
+
```Python
|
| 189 |
+
from typing import Literal
|
| 190 |
+
|
| 191 |
+
from pydantic import Field, BaseModel
|
| 192 |
+
from numind.nuextract_utils import convert_json_schema_to_nuextract_template
|
| 193 |
+
|
| 194 |
+
|
| 195 |
+
class HotelBooking(BaseModel):
|
| 196 |
+
city: str
|
| 197 |
+
check_in_date: str = Field(description="date")
|
| 198 |
+
check_out_date: str = Field(description="date")
|
| 199 |
+
number_of_guests: int
|
| 200 |
+
room_type: Literal["single", "double", "suite"]
|
| 201 |
+
|
| 202 |
+
|
| 203 |
+
template, dropped_branches = convert_json_schema_to_nuextract_template(
|
| 204 |
+
HotelBooking.model_json_schema()
|
| 205 |
+
)
|
| 206 |
+
|
| 207 |
+
# {'check_in_date': 'date', 'check_out_date': 'date', 'city': 'string', 'number_of_guests': 'integer', 'room_type': ['single', 'double', 'suite']}
|
| 208 |
+
```
|
| 209 |
+
|
| 210 |
+
## Document-to-Markdown
|
| 211 |
+
|
| 212 |
+
NuExtract can also convert document images into clean Markdown. Output will be markdown for text (headers etc), html for tables, latex for mat and ```<figure data-type="image" data-id="img_n"><img src="/NM-dev/model_card-A/resolve/main/img_n.png" alt="Detail description of the images"/> ```
|
| 213 |
+
|
| 214 |
+
Markdown example:
|
| 215 |
+
|
| 216 |
+
```markdown
|
| 217 |
+
<figure data-type="image" data-id="img_1">
|
| 218 |
+
<img src="img_1.png" alt="Logo of Mobilier 2000 with contact information: Tél.: (418) 275-4232, 1654, boul. Marcotte, Roberval (Qc) G8H 2P2"/>
|
| 219 |
+
</figure>
|
| 220 |
+
|
| 221 |
+
# COMMANDE
|
| 222 |
+
**NUMÉRO 72259**
|
| 223 |
+
|
| 224 |
+
1
|
| 225 |
+
|
| 226 |
+
**Vendu à**
|
| 227 |
+
TREMBLAY ERIC
|
| 228 |
+
ERIC TREMBLAY
|
| 229 |
+
348 BOUL. DE L'ANSE
|
| 230 |
+
ROBERVAL
|
| 231 |
+
G8H 1Y9
|
| 232 |
+
|
| 233 |
+
**Livré à**
|
| 234 |
+
TREMBLAY ERIC
|
| 235 |
+
ERIC TREMBLAY
|
| 236 |
+
348 BOUL. DE L'ANSE
|
| 237 |
+
ROBERVAL
|
| 238 |
+
G8H 1Y9
|
| 239 |
+
|
| 240 |
+
<table>
|
| 241 |
+
<thead>
|
| 242 |
+
<tr>
|
| 243 |
+
<th># CLIENT</th>
|
| 244 |
+
<th>EXPÉDITEUR</th>
|
| 245 |
+
<th>TERME DE CRÉDIT</th>
|
| 246 |
+
<th>DATE</th>
|
| 247 |
+
</tr>
|
| 248 |
+
</thead>
|
| 249 |
+
<tbody>
|
| 250 |
+
<tr>
|
| 251 |
+
<td>2753133</td>
|
| 252 |
+
<td>Notre camion</td>
|
| 253 |
+
<td>à la livraison</td>
|
| 254 |
+
<td>22/06/2023</td>
|
| 255 |
+
</tr>
|
| 256 |
+
</tbody>
|
| 257 |
+
</table>
|
| 258 |
+
|
| 259 |
+
<table>
|
| 260 |
+
<thead>
|
| 261 |
+
<tr>
|
| 262 |
+
<th>NOM DU VENDEUR</th>
|
| 263 |
+
<th>VOTRE ÉCONOMIE !</th>
|
| 264 |
+
<th># COMMANDE</th>
|
| 265 |
+
</tr>
|
| 266 |
+
</thead>
|
| 267 |
+
<tbody>
|
| 268 |
+
<tr>
|
| 269 |
+
<td>Éric</td>
|
| 270 |
+
<td>0.00</td>
|
| 271 |
+
<td></td>
|
| 272 |
+
</tr>
|
| 273 |
+
</tbody>
|
| 274 |
+
</table>
|
| 275 |
+
```
|
| 276 |
+
|
| 277 |
+
---
|
| 278 |
+
|
| 279 |
+
## Reasoning and non-reasoning modes
|
| 280 |
+
|
| 281 |
+
NuExtract supports both reasoning and non-reasoning inference.
|
| 282 |
+
|
| 283 |
+
### Non-thinking mode
|
| 284 |
+
|
| 285 |
+
Use this for fast and deterministic extraction or Markdown conversion.
|
| 286 |
+
|
| 287 |
+
```python
|
| 288 |
+
enable_thinking = False
|
| 289 |
+
temperature = 0.2
|
| 290 |
+
```
|
| 291 |
+
|
| 292 |
+
### Thinking mode
|
| 293 |
+
|
| 294 |
+
Use this for difficult documents, complex layouts, ambiguous fields, or cases where the document structure requires additional reasoning.
|
| 295 |
+
|
| 296 |
+
```python
|
| 297 |
+
enable_thinking = True
|
| 298 |
+
temperature = 0.6
|
| 299 |
+
```
|
| 300 |
+
|
| 301 |
+
For production extraction workloads, we recommend starting with **non-reasoning mode** and enabling reasoning only for difficult examples.
|
| 302 |
+
|
| 303 |
+
|
| 304 |
+
---
|
| 305 |
+
|
| 306 |
+
## vLLM deployment
|
| 307 |
+
|
| 308 |
+
NuExtract can be served with vLLM using an OpenAI-compatible API.
|
| 309 |
+
|
| 310 |
+
```bash
|
| 311 |
+
vllm serve numind/NuExtract3 \
|
| 312 |
+
--trust-remote-code \
|
| 313 |
+
--limit-mm-per-prompt '{"image": 99, "video": 0}' \
|
| 314 |
+
--chat-template-content-format openai \
|
| 315 |
+
--generation-config vllm \
|
| 316 |
+
--max-model-len 131072 \
|
| 317 |
+
--speculative-config '{"method": "qwen3_next_mtp", "num_speculative_tokens": 2}'
|
| 318 |
+
```
|
| 319 |
+
|
| 320 |
+
|
| 321 |
+
### Multi Token Prediction
|
| 322 |
+
<details>
|
| 323 |
+
The deployment commands above enable Multi Token Prediction (MTP) through vLLM speculative decoding:
|
| 324 |
+
|
| 325 |
+
```bash
|
| 326 |
+
--speculative-config '{"method": "qwen3_next_mtp", "num_speculative_tokens": 2}'
|
| 327 |
+
```
|
| 328 |
+
|
| 329 |
+
MTP can improve decoding throughput without changing the OpenAI-compatible request payload. You can tune `num_speculative_tokens` for your hardware and workload, or remove `--speculative-config` if your vLLM version or environment does not support this speculative decoding method.
|
| 330 |
+
|
| 331 |
+
If you encounter memory issues, reduce the maximum model length and the maximum number of images:
|
| 332 |
+
|
| 333 |
+
```bash
|
| 334 |
+
vllm serve numind/NuExtract-3 \
|
| 335 |
+
--trust-remote-code \
|
| 336 |
+
--limit-mm-per-prompt '{"image": 6, "video": 0}' \
|
| 337 |
+
--chat-template-content-format openai \
|
| 338 |
+
--generation-config vllm \
|
| 339 |
+
--max-model-len 16384 \
|
| 340 |
+
--speculative-config '{"method": "qwen3_next_mtp", "num_speculative_tokens": 2}'
|
| 341 |
+
```
|
| 342 |
+
</details>
|
| 343 |
+
|
| 344 |
+
## vLLM inference: structured extraction: text
|
| 345 |
+
```python
|
| 346 |
+
import json
|
| 347 |
+
from openai import OpenAI
|
| 348 |
+
|
| 349 |
+
client = OpenAI(
|
| 350 |
+
api_key="EMPTY",
|
| 351 |
+
base_url="http://localhost:8000/v1",
|
| 352 |
+
)
|
| 353 |
+
|
| 354 |
+
template = {
|
| 355 |
+
"store": "verbatim-string",
|
| 356 |
+
"date": "date-time",
|
| 357 |
+
"total": "number",
|
| 358 |
+
"currency": ["USD", "EUR", "GBP", "JPY", "Other"],
|
| 359 |
+
"items": [
|
| 360 |
+
{
|
| 361 |
+
"name": "verbatim-string",
|
| 362 |
+
"price": "number"
|
| 363 |
+
}
|
| 364 |
+
]
|
| 365 |
+
}
|
| 366 |
+
|
| 367 |
+
response = client.chat.completions.create(
|
| 368 |
+
model="numind/NuExtract3",
|
| 369 |
+
temperature=0.2,
|
| 370 |
+
messages=[
|
| 371 |
+
{
|
| 372 |
+
"role": "user",
|
| 373 |
+
"content": [
|
| 374 |
+
{
|
| 375 |
+
"type": "text",
|
| 376 |
+
"text": "Yesterday I bought apples and coffee at Trader Joe's for a total of $12.40."
|
| 377 |
+
}
|
| 378 |
+
],
|
| 379 |
+
}
|
| 380 |
+
],
|
| 381 |
+
extra_body={
|
| 382 |
+
"chat_template_kwargs": {
|
| 383 |
+
"template": json.dumps(template, indent=4),
|
| 384 |
+
"enable_thinking": False
|
| 385 |
+
}
|
| 386 |
+
}
|
| 387 |
+
)
|
| 388 |
+
|
| 389 |
+
print(response.choices[0].message.content)
|
| 390 |
+
```
|
| 391 |
+
|
| 392 |
+
Example output:
|
| 393 |
+
|
| 394 |
+
```json
|
| 395 |
+
{
|
| 396 |
+
"store": "Trader Joe's",
|
| 397 |
+
"date": null,
|
| 398 |
+
"total": 12.40,
|
| 399 |
+
"currency": "USD",
|
| 400 |
+
"items": [
|
| 401 |
+
{
|
| 402 |
+
"name": "apples",
|
| 403 |
+
"price": null
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"name": "coffee",
|
| 407 |
+
"price": null
|
| 408 |
+
}
|
| 409 |
+
]
|
| 410 |
+
}
|
| 411 |
+
```
|
| 412 |
+
|
| 413 |
+
---
|
| 414 |
+
|
| 415 |
+
## vLLM inference: structured extraction: image
|
| 416 |
+
|
| 417 |
+
```python
|
| 418 |
+
import json
|
| 419 |
+
import base64
|
| 420 |
+
from openai import OpenAI
|
| 421 |
+
|
| 422 |
+
client = OpenAI(
|
| 423 |
+
api_key="EMPTY",
|
| 424 |
+
base_url="http://localhost:8000/v1",
|
| 425 |
+
)
|
| 426 |
+
|
| 427 |
+
def encode_image(image_path):
|
| 428 |
+
with open(image_path, "rb") as image_file:
|
| 429 |
+
return base64.b64encode(image_file.read()).decode("utf-8")
|
| 430 |
+
|
| 431 |
+
image_base64 = encode_image("receipt.png")
|
| 432 |
+
data_url = f"data:image/png;base64,{image_base64}"
|
| 433 |
+
|
| 434 |
+
template = {
|
| 435 |
+
"store": "verbatim-string",
|
| 436 |
+
"date": "date-time",
|
| 437 |
+
"total": "number",
|
| 438 |
+
"payment_method": "verbatim-string"
|
| 439 |
+
}
|
| 440 |
+
|
| 441 |
+
response = client.chat.completions.create(
|
| 442 |
+
model="numind/NuExtract3",
|
| 443 |
+
temperature=0.2,
|
| 444 |
+
messages=[
|
| 445 |
+
{
|
| 446 |
+
"role": "user",
|
| 447 |
+
"content": [
|
| 448 |
+
{
|
| 449 |
+
"type": "image_url",
|
| 450 |
+
"image_url": {"url": data_url}
|
| 451 |
+
}
|
| 452 |
+
],
|
| 453 |
+
}
|
| 454 |
+
],
|
| 455 |
+
extra_body={
|
| 456 |
+
"chat_template_kwargs": {
|
| 457 |
+
"template": json.dumps(template, indent=4),
|
| 458 |
+
"enable_thinking": False
|
| 459 |
+
}
|
| 460 |
+
}
|
| 461 |
+
)
|
| 462 |
+
|
| 463 |
+
print(response.choices[0].message.content)
|
| 464 |
+
```
|
| 465 |
+
|
| 466 |
+
Example output:
|
| 467 |
+
|
| 468 |
+
```json
|
| 469 |
+
{
|
| 470 |
+
"store": "Trader Joe's",
|
| 471 |
+
"date": "2025-04-12",
|
| 472 |
+
"total": 42.85,
|
| 473 |
+
"payment_method": "Visa"
|
| 474 |
+
}
|
| 475 |
+
```
|
| 476 |
+
|
| 477 |
+
### Multiple page PDF
|
| 478 |
+
<details>
|
| 479 |
+
You can render a PDF to one PNG image per page with PyMuPDF, then pass the images to vLLM in page order.
|
| 480 |
+
|
| 481 |
+
```python
|
| 482 |
+
import base64
|
| 483 |
+
import json
|
| 484 |
+
|
| 485 |
+
import fitz # pip install pymupdf
|
| 486 |
+
from openai import OpenAI
|
| 487 |
+
|
| 488 |
+
client = OpenAI(
|
| 489 |
+
api_key="EMPTY",
|
| 490 |
+
base_url="http://localhost:8000/v1",
|
| 491 |
+
)
|
| 492 |
+
|
| 493 |
+
def pdf_to_png_data_urls(pdf_path, dpi=170):
|
| 494 |
+
data_urls = []
|
| 495 |
+
|
| 496 |
+
with fitz.open(pdf_path) as doc:
|
| 497 |
+
for page in doc:
|
| 498 |
+
pix = page.get_pixmap(dpi=dpi, alpha=False)
|
| 499 |
+
png_bytes = pix.tobytes("png")
|
| 500 |
+
png_base64 = base64.b64encode(png_bytes).decode("utf-8")
|
| 501 |
+
data_urls.append(f"data:image/png;base64,{png_base64}")
|
| 502 |
+
|
| 503 |
+
return data_urls
|
| 504 |
+
|
| 505 |
+
data_urls = pdf_to_png_data_urls("invoice.pdf", dpi=170)
|
| 506 |
+
|
| 507 |
+
template = {
|
| 508 |
+
"invoice_number": "verbatim-string",
|
| 509 |
+
"invoice_date": "date",
|
| 510 |
+
"total": "number",
|
| 511 |
+
"currency": "currency",
|
| 512 |
+
"line_items": [
|
| 513 |
+
{
|
| 514 |
+
"description": "verbatim-string",
|
| 515 |
+
"quantity": "number",
|
| 516 |
+
"unit_price": "number",
|
| 517 |
+
"total": "number"
|
| 518 |
+
}
|
| 519 |
+
]
|
| 520 |
+
}
|
| 521 |
+
|
| 522 |
+
response = client.chat.completions.create(
|
| 523 |
+
model="numind/NuExtract3",
|
| 524 |
+
temperature=0.2,
|
| 525 |
+
messages=[
|
| 526 |
+
{
|
| 527 |
+
"role": "user",
|
| 528 |
+
"content": [
|
| 529 |
+
{
|
| 530 |
+
"type": "image_url",
|
| 531 |
+
"image_url": {"url": data_url}
|
| 532 |
+
}
|
| 533 |
+
for data_url in data_urls
|
| 534 |
+
],
|
| 535 |
+
}
|
| 536 |
+
],
|
| 537 |
+
extra_body={
|
| 538 |
+
"chat_template_kwargs": {
|
| 539 |
+
"template": json.dumps(template, indent=4),
|
| 540 |
+
"enable_thinking": False
|
| 541 |
+
}
|
| 542 |
+
}
|
| 543 |
+
)
|
| 544 |
+
|
| 545 |
+
print(response.choices[0].message.content)
|
| 546 |
+
```
|
| 547 |
+
</details>
|
| 548 |
+
|
| 549 |
+
|
| 550 |
+
|
| 551 |
+
## vLLM inference: document-to-Markdown
|
| 552 |
+
|
| 553 |
+
For Markdown OCR, use `mode="markdown"` or `mode="content"` without a template.
|
| 554 |
+
|
| 555 |
+
```python
|
| 556 |
+
import base64
|
| 557 |
+
from openai import OpenAI
|
| 558 |
+
|
| 559 |
+
client = OpenAI(
|
| 560 |
+
api_key="EMPTY",
|
| 561 |
+
base_url="http://localhost:8000/v1",
|
| 562 |
+
)
|
| 563 |
+
|
| 564 |
+
def encode_image(image_path):
|
| 565 |
+
with open(image_path, "rb") as image_file:
|
| 566 |
+
return base64.b64encode(image_file.read()).decode("utf-8")
|
| 567 |
+
|
| 568 |
+
image_base64 = encode_image("document.png")
|
| 569 |
+
data_url = f"data:image/png;base64,{image_base64}"
|
| 570 |
+
|
| 571 |
+
response = client.chat.completions.create(
|
| 572 |
+
model="numind/NuExtract3",
|
| 573 |
+
temperature=0,
|
| 574 |
+
messages=[
|
| 575 |
+
{
|
| 576 |
+
"role": "user",
|
| 577 |
+
"content": [
|
| 578 |
+
{
|
| 579 |
+
"type": "image_url",
|
| 580 |
+
"image_url": {"url": data_url}
|
| 581 |
+
}
|
| 582 |
+
],
|
| 583 |
+
}
|
| 584 |
+
],
|
| 585 |
+
extra_body={
|
| 586 |
+
"chat_template_kwargs": {
|
| 587 |
+
"mode": "markdown",
|
| 588 |
+
"enable_thinking": False
|
| 589 |
+
}
|
| 590 |
+
}
|
| 591 |
+
)
|
| 592 |
+
|
| 593 |
+
print(response.choices[0].message.content)
|
| 594 |
+
```
|
| 595 |
+
|
| 596 |
+
---
|
| 597 |
+
|
| 598 |
+
## vLLM inference: reasoning mode
|
| 599 |
+
<details>
|
| 600 |
+
Reasoning can be enabled for harder structured extraction or Markdown tasks.
|
| 601 |
+
|
| 602 |
+
```python
|
| 603 |
+
response = client.chat.completions.create(
|
| 604 |
+
model="numind/NuExtract3",
|
| 605 |
+
temperature=0.7,
|
| 606 |
+
messages=[
|
| 607 |
+
{
|
| 608 |
+
"role": "user",
|
| 609 |
+
"content": [
|
| 610 |
+
{
|
| 611 |
+
"type": "image_url",
|
| 612 |
+
"image_url": {"url": data_url}
|
| 613 |
+
}
|
| 614 |
+
],
|
| 615 |
+
}
|
| 616 |
+
],
|
| 617 |
+
extra_body={
|
| 618 |
+
"chat_template_kwargs": {
|
| 619 |
+
"mode": "markdown",
|
| 620 |
+
"enable_thinking": True
|
| 621 |
+
}
|
| 622 |
+
}
|
| 623 |
+
)
|
| 624 |
+
|
| 625 |
+
result = response.choices[0].message.content
|
| 626 |
+
|
| 627 |
+
if "</think>" in result:
|
| 628 |
+
reasoning = result.split("<think>")[1].split("</think>")[0]
|
| 629 |
+
answer = result.split("</think>")[-1].strip()
|
| 630 |
+
else:
|
| 631 |
+
reasoning = None
|
| 632 |
+
answer = result
|
| 633 |
+
|
| 634 |
+
print(answer)
|
| 635 |
+
```
|
| 636 |
+
</details>
|
| 637 |
+
|
| 638 |
+
|
| 639 |
+
## In-context examples for extraction
|
| 640 |
+
<details>
|
| 641 |
+
NuExtract supports in-context examples for structured extraction.
|
| 642 |
+
|
| 643 |
+
Examples are especially useful when the desired formatting is ambiguous or when the schema requires task-specific conventions. Examples can be provided by using `developer` messages, for which all items of the contents except the last one are the input, and the last one is the expected output.
|
| 644 |
+
|
| 645 |
+
```python
|
| 646 |
+
import json
|
| 647 |
+
from openai import OpenAI
|
| 648 |
+
|
| 649 |
+
client = OpenAI(
|
| 650 |
+
api_key="EMPTY",
|
| 651 |
+
base_url="http://localhost:8000/v1",
|
| 652 |
+
)
|
| 653 |
+
|
| 654 |
+
template = {
|
| 655 |
+
"names": ["string"]
|
| 656 |
+
}
|
| 657 |
+
|
| 658 |
+
response = client.chat.completions.create(
|
| 659 |
+
model="numind/NuExtract3",
|
| 660 |
+
temperature=0,
|
| 661 |
+
messages=[
|
| 662 |
+
{
|
| 663 |
+
"role": "developer",
|
| 664 |
+
"content": [
|
| 665 |
+
{
|
| 666 |
+
"type": "text",
|
| 667 |
+
"text": "Stephen is the manager at Susan's store.",
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"type": "text",
|
| 671 |
+
"text": "{\"names\": [\"-STEPHEN-\", \"-SUSAN-\"]}",
|
| 672 |
+
}
|
| 673 |
+
],
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"role": "user",
|
| 677 |
+
"content": [
|
| 678 |
+
{
|
| 679 |
+
"type": "text",
|
| 680 |
+
"text": "John went to the restaurant with Mary. James went to the cinema."
|
| 681 |
+
}
|
| 682 |
+
],
|
| 683 |
+
}
|
| 684 |
+
],
|
| 685 |
+
extra_body={
|
| 686 |
+
"chat_template_kwargs": {
|
| 687 |
+
"template": json.dumps(template, indent=4),
|
| 688 |
+
"enable_thinking": False
|
| 689 |
+
}
|
| 690 |
+
}
|
| 691 |
+
)
|
| 692 |
+
|
| 693 |
+
print(response.choices[0].message.content)
|
| 694 |
+
```
|
| 695 |
+
|
| 696 |
+
Example output:
|
| 697 |
+
|
| 698 |
+
```json
|
| 699 |
+
{
|
| 700 |
+
"names": ["-JOHN-", "-MARY-", "-JAMES-"]
|
| 701 |
+
}
|
| 702 |
+
```
|
| 703 |
+
</details>
|
| 704 |
+
|
| 705 |
+
|
| 706 |
+
## vLLM inference: template generation
|
| 707 |
+
|
| 708 |
+
NuExtract can generate an extraction template from a natural language description.
|
| 709 |
+
|
| 710 |
+
```python
|
| 711 |
+
from openai import OpenAI
|
| 712 |
+
|
| 713 |
+
client = OpenAI(
|
| 714 |
+
api_key="EMPTY",
|
| 715 |
+
base_url="http://localhost:8000/v1",
|
| 716 |
+
)
|
| 717 |
+
|
| 718 |
+
response = client.chat.completions.create(
|
| 719 |
+
model="numind/NuExtract3",
|
| 720 |
+
temperature=0,
|
| 721 |
+
messages=[
|
| 722 |
+
{
|
| 723 |
+
"role": "user",
|
| 724 |
+
"content": [
|
| 725 |
+
{
|
| 726 |
+
"type": "text",
|
| 727 |
+
"text": "I want to extract the key details from a rental contract."
|
| 728 |
+
}
|
| 729 |
+
],
|
| 730 |
+
}
|
| 731 |
+
],
|
| 732 |
+
extra_body={
|
| 733 |
+
"chat_template_kwargs": {
|
| 734 |
+
"mode": "template-generation"
|
| 735 |
+
}
|
| 736 |
+
}
|
| 737 |
+
)
|
| 738 |
+
|
| 739 |
+
print(response.choices[0].message.content)
|
| 740 |
+
```
|
| 741 |
+
|
| 742 |
+
Example output:
|
| 743 |
+
|
| 744 |
+
```json
|
| 745 |
+
{
|
| 746 |
+
"contract_title": "verbatim-string",
|
| 747 |
+
"landlord": "verbatim-string",
|
| 748 |
+
"tenant": "verbatim-string",
|
| 749 |
+
"property_address": "verbatim-string",
|
| 750 |
+
"start_date": "date-time",
|
| 751 |
+
"end_date": "date-time",
|
| 752 |
+
"monthly_rent": "number",
|
| 753 |
+
"currency": "verbatim-string",
|
| 754 |
+
"deposit": "number",
|
| 755 |
+
"signatories": ["verbatim-string"]
|
| 756 |
+
}
|
| 757 |
+
```
|
| 758 |
+
|
| 759 |
+
## Curl examples
|
| 760 |
+
<details>
|
| 761 |
+
|
| 762 |
+
The following examples assume that vLLM is running locally on port 8000. They use `jq` to build valid JSON request bodies without manually escaping the image data or template string.
|
| 763 |
+
|
| 764 |
+
### Single image structured extraction
|
| 765 |
+
|
| 766 |
+
```bash
|
| 767 |
+
API_KEY="EMPTY"
|
| 768 |
+
IMAGE_BASE64_FILE=$(mktemp)
|
| 769 |
+
REQUEST_BODY_FILE=$(mktemp)
|
| 770 |
+
|
| 771 |
+
base64 < receipt.png | tr -d '\n' > "$IMAGE_BASE64_FILE"
|
| 772 |
+
|
| 773 |
+
TEMPLATE=$(cat <<'JSON'
|
| 774 |
+
{
|
| 775 |
+
"store": "verbatim-string",
|
| 776 |
+
"date": "date-time",
|
| 777 |
+
"total": "number",
|
| 778 |
+
"payment_method": "verbatim-string"
|
| 779 |
+
}
|
| 780 |
+
JSON
|
| 781 |
+
)
|
| 782 |
+
|
| 783 |
+
jq -n \
|
| 784 |
+
--rawfile image_base64 "$IMAGE_BASE64_FILE" \
|
| 785 |
+
--arg template "$TEMPLATE" \
|
| 786 |
+
'{
|
| 787 |
+
model: "numind/NuExtract3",
|
| 788 |
+
temperature: 0,
|
| 789 |
+
messages: [
|
| 790 |
+
{
|
| 791 |
+
role: "user",
|
| 792 |
+
content: [
|
| 793 |
+
{
|
| 794 |
+
type: "image_url",
|
| 795 |
+
image_url: {url: ("data:image/png;base64," + $image_base64)}
|
| 796 |
+
}
|
| 797 |
+
]
|
| 798 |
+
}
|
| 799 |
+
],
|
| 800 |
+
chat_template_kwargs: {
|
| 801 |
+
template: $template,
|
| 802 |
+
enable_thinking: false
|
| 803 |
+
}
|
| 804 |
+
}' > "$REQUEST_BODY_FILE"
|
| 805 |
+
|
| 806 |
+
curl http://localhost:8000/v1/chat/completions \
|
| 807 |
+
-H "Content-Type: application/json" \
|
| 808 |
+
-H "Authorization: Bearer $API_KEY" \
|
| 809 |
+
--data-binary "@$REQUEST_BODY_FILE"
|
| 810 |
+
|
| 811 |
+
rm "$IMAGE_BASE64_FILE" "$REQUEST_BODY_FILE"
|
| 812 |
+
```
|
| 813 |
+
|
| 814 |
+
### Single image content extraction
|
| 815 |
+
|
| 816 |
+
```bash
|
| 817 |
+
API_KEY="EMPTY"
|
| 818 |
+
IMAGE_BASE64_FILE=$(mktemp)
|
| 819 |
+
REQUEST_BODY_FILE=$(mktemp)
|
| 820 |
+
|
| 821 |
+
base64 < document.png | tr -d '\n' > "$IMAGE_BASE64_FILE"
|
| 822 |
+
|
| 823 |
+
jq -n \
|
| 824 |
+
--rawfile image_base64 "$IMAGE_BASE64_FILE" \
|
| 825 |
+
'{
|
| 826 |
+
model: "numind/NuExtract3",
|
| 827 |
+
temperature: 0,
|
| 828 |
+
messages: [
|
| 829 |
+
{
|
| 830 |
+
role: "user",
|
| 831 |
+
content: [
|
| 832 |
+
{
|
| 833 |
+
type: "image_url",
|
| 834 |
+
image_url: {url: ("data:image/png;base64," + $image_base64)}
|
| 835 |
+
}
|
| 836 |
+
]
|
| 837 |
+
}
|
| 838 |
+
],
|
| 839 |
+
chat_template_kwargs: {
|
| 840 |
+
mode: "content",
|
| 841 |
+
enable_thinking: false
|
| 842 |
+
}
|
| 843 |
+
}' > "$REQUEST_BODY_FILE"
|
| 844 |
+
|
| 845 |
+
curl http://localhost:8000/v1/chat/completions \
|
| 846 |
+
-H "Content-Type: application/json" \
|
| 847 |
+
-H "Authorization: Bearer $API_KEY" \
|
| 848 |
+
--data-binary "@$REQUEST_BODY_FILE"
|
| 849 |
+
|
| 850 |
+
rm "$IMAGE_BASE64_FILE" "$REQUEST_BODY_FILE"
|
| 851 |
+
```
|
| 852 |
+
</details>
|
| 853 |
+
|
| 854 |
+
|
| 855 |
+
## Transformers example
|
| 856 |
+
<details>
|
| 857 |
+
You can also run NuExtract directly with `transformers`. The same `template`, `mode`, and `enable_thinking` options are passed to `processor.apply_chat_template`.
|
| 858 |
+
|
| 859 |
+
```python
|
| 860 |
+
import json
|
| 861 |
+
|
| 862 |
+
import torch
|
| 863 |
+
from PIL import Image
|
| 864 |
+
from transformers import AutoModelForImageTextToText, AutoProcessor
|
| 865 |
+
|
| 866 |
+
model_id = "numind/NuExtract3"
|
| 867 |
+
|
| 868 |
+
processor = AutoProcessor.from_pretrained(
|
| 869 |
+
model_id,
|
| 870 |
+
trust_remote_code=True,
|
| 871 |
+
)
|
| 872 |
+
model = AutoModelForImageTextToText.from_pretrained(
|
| 873 |
+
model_id,
|
| 874 |
+
dtype=torch.bfloat16,
|
| 875 |
+
device_map="auto",
|
| 876 |
+
trust_remote_code=True,
|
| 877 |
+
).eval()
|
| 878 |
+
|
| 879 |
+
def run_nuextract(messages, **chat_template_kwargs):
|
| 880 |
+
inputs = processor.apply_chat_template(
|
| 881 |
+
messages,
|
| 882 |
+
add_generation_prompt=True,
|
| 883 |
+
tokenize=True,
|
| 884 |
+
return_dict=True,
|
| 885 |
+
return_tensors="pt",
|
| 886 |
+
**chat_template_kwargs,
|
| 887 |
+
).to(model.device)
|
| 888 |
+
|
| 889 |
+
with torch.inference_mode():
|
| 890 |
+
generated_ids = model.generate(
|
| 891 |
+
**inputs,
|
| 892 |
+
max_new_tokens=4096,
|
| 893 |
+
do_sample=False,
|
| 894 |
+
)
|
| 895 |
+
|
| 896 |
+
generated_ids = generated_ids[:, inputs.input_ids.shape[1]:]
|
| 897 |
+
return processor.batch_decode(
|
| 898 |
+
generated_ids,
|
| 899 |
+
skip_special_tokens=True,
|
| 900 |
+
clean_up_tokenization_spaces=False,
|
| 901 |
+
)[0].strip()
|
| 902 |
+
|
| 903 |
+
# Single image structured extraction
|
| 904 |
+
receipt_image = Image.open("receipt.png").convert("RGB")
|
| 905 |
+
receipt_messages = [
|
| 906 |
+
{
|
| 907 |
+
"role": "user",
|
| 908 |
+
"content": [
|
| 909 |
+
{
|
| 910 |
+
"type": "image",
|
| 911 |
+
"image": receipt_image,
|
| 912 |
+
}
|
| 913 |
+
],
|
| 914 |
+
}
|
| 915 |
+
]
|
| 916 |
+
|
| 917 |
+
template = {
|
| 918 |
+
"store": "verbatim-string",
|
| 919 |
+
"date": "date-time",
|
| 920 |
+
"total": "number",
|
| 921 |
+
"payment_method": "verbatim-string"
|
| 922 |
+
}
|
| 923 |
+
|
| 924 |
+
structured_output = run_nuextract(
|
| 925 |
+
receipt_messages,
|
| 926 |
+
template=json.dumps(template, indent=4),
|
| 927 |
+
enable_thinking=False,
|
| 928 |
+
)
|
| 929 |
+
print(structured_output)
|
| 930 |
+
|
| 931 |
+
# Single image content extraction
|
| 932 |
+
document_image = Image.open("document.png").convert("RGB")
|
| 933 |
+
document_messages = [
|
| 934 |
+
{
|
| 935 |
+
"role": "user",
|
| 936 |
+
"content": [
|
| 937 |
+
{
|
| 938 |
+
"type": "image",
|
| 939 |
+
"image": document_image,
|
| 940 |
+
}
|
| 941 |
+
],
|
| 942 |
+
}
|
| 943 |
+
]
|
| 944 |
+
|
| 945 |
+
content_output = run_nuextract(
|
| 946 |
+
document_messages,
|
| 947 |
+
mode="content",
|
| 948 |
+
enable_thinking=False,
|
| 949 |
+
)
|
| 950 |
+
print(content_output)
|
| 951 |
+
```
|
| 952 |
+
</details>
|
| 953 |
+
|
| 954 |
+
Special thanks to the Lambda.ai team for the compute that made this project a success.
|
| 955 |
+
|
| 956 |
+
## Citation
|
| 957 |
+
|
| 958 |
+
If you use NuExtract, please cite NuMind and link to the model page.
|
| 959 |
+
|
| 960 |
+
```bibtex
|
| 961 |
+
@misc{nuextract3,
|
| 962 |
+
title = {NuExtract3},
|
| 963 |
+
author = {NuMind},
|
| 964 |
+
year = {2026},
|
| 965 |
+
url = {https://nuextract.ai/}
|
| 966 |
+
}
|
| 967 |
+
```
|
TYPES.md
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
| Type | Description | Examples |
|
| 2 |
+
| --- | --- | --- |
|
| 3 |
+
| **integer** | An integer number. | 12, 0, -4 |
|
| 4 |
+
| **number** | Any number, including floating point or integers. | 3.14, -9.1, 0 |
|
| 5 |
+
| **string** | A general string; can be abstractive or deduced from reasoning. | Hello World, any string |
|
| 6 |
+
| **verbatim-string** | Strictly extractive from input; preserves all characters (accents, emojis) but normalizes whitespace/tabs to a single space. | John Doe, 1120 Santa Monica Boulevard |
|
| 7 |
+
| **date** | ISO 8601 compliant. Supports reduced accuracy (YYYY-MM, YYYY, --MM-DD) and week dates (YYYY-Www). | 2024-01-15, 2024-01, --12-25 |
|
| 8 |
+
| **time** | ISO 8601 compliant. Supports reduced accuracy and timezone offsets (+hh-mm). | 14:30:57, 18:01, 14:30:45.123Z |
|
| 9 |
+
| **date-time** | ISO 8601 compliant (YYYY-MM-DDThh:mm:ss.s+hh-mm). Can omit components if only date or time is present. | 2024-03-14T14:45:00, 2023-05-15T14 |
|
| 10 |
+
| **duration** | ISO 8601 duration (PnYnMnDTnHnMnS). "P3W" (weeks) cannot be combined with other date components. | P2Y1M3D, PT1M30S, P3W |
|
| 11 |
+
| **boolean** | A logic value of true or false. | true, false |
|
| 12 |
+
| **country** | Uppercase 2-character ISO 3166-1 country code. | FR, SG, KR |
|
| 13 |
+
| **currency** | Uppercase 3-character ISO 4217 code. Covers current and historic currencies. | EUR, USD, DEM |
|
| 14 |
+
| **language** | Lowercase 3-character ISO 639-3 language code. | eng, fra, cos |
|
| 15 |
+
| **language-tag** | IETF BCP 47 / RFC 5646 tag. Includes language, script (opt), region (opt), and variants. | en-US, zh-Hans-CN, sl-rozaj |
|
| 16 |
+
| **script** | Titlecase 4-character ISO 15924 script code. | Latn, Kore, Deva |
|
| 17 |
+
| **url** | RFC 3987 IRI. Supports Unicode characters, schemes (http, ftp), and Punycode for domain names. | https://例子.测试/路径, ftp://user@host/file.txt |
|
| 18 |
+
| **email-address** | RFC 5322/6531 compliant. Supports internationalized characters in local and domain parts. | firstname.lastname@example.com, 用户@例子.公司 |
|
| 19 |
+
| **phone-number** | E.164 compliant if region is known (e.g., +1...); otherwise, extracted as a raw digit string. | +33612345678, 6505550123 |
|
| 20 |
+
| **iban** | ISO 13616-1 International Bank Account Number. Structure varies by country. | DE89370400440532013000 |
|
| 21 |
+
| **bic** | ISO 9362 Business Identifier Code (8 or 11 characters). | BNPAFRPPXXX, DEUTDEDBFRA |
|
| 22 |
+
| **unit-code** | UCUM (Unified Code for Units of Measure) code. | m, kg, s, Hz |
|
| 23 |
+
| **region:US** | Uppercase subdivision code complying to ISO 3166-2:US. | NY, DC, GU |
|
| 24 |
+
| **region:FR** | Uppercase subdivision code complying to ISO 3166-2:FR. | 49 (Maine-et-Loire), MQ (Martinique), V (Rhône-Alpes) |
|
| 25 |
+
| **region:IE** | Uppercase subdivision code complying to ISO 3166-2:IE. | D (Dublin), C (Connacht), WD (Waterford) |
|
| 26 |
+
| **region:GB** | Uppercase subdivision code complying to ISO 3166-2:GB. | WSX (West Sussex), WSM (Westminster), WIL (Wiltshire) |
|
| 27 |
+
| **region:IT** | Uppercase subdivision code complying to ISO 3166-2:IT. | RM (Rome), BZ (Bolzano), 82 (Sicily) |
|
| 28 |
+
| **region:ES** | Uppercase subdivision code complying to ISO 3166-2:ES. | GA (Galicia), GR (Granada), ML (Melilla) |
|
| 29 |
+
| **region:DE** | Uppercase subdivision code complying to ISO 3166-2:DE. | BY (Bayern), BE (Berlin), HH (Hamburg) |
|
| 30 |
+
| **region:PT** | Uppercase subdivision code complying to ISO 3166-2:PT. | 11 (Lisbon), 20 (Azores) |
|
| 31 |
+
| **region:CA** | Uppercase subdivision code complying to ISO 3166-2:CA. | QC (Quebec), NU (Nunavut), YT (Yukon) |
|
| 32 |
+
| **region:MX** | Uppercase subdivision code complying to ISO 3166-2:MX. | JAL (Jalisco), DIF (Distrito Federal), AGU (Aguascalientes) |
|
| 33 |
+
| **region:BR** | Uppercase subdivision code complying to ISO 3166-2:BR. | RJ (Rio de Janeiro), DF (Distrito Federal), SP (São Paulo) |
|
| 34 |
+
| **region:AU** | Uppercase subdivision code complying to ISO 3166-2:AU. | NSW (New South Wales), VIC (Victoria), ACT (Australian Capital Territory) |
|
| 35 |
+
| **region:JP** | Uppercase subdivision code complying to ISO 3166-2:JP. | 13 (Tokyo), 27 (Osaka), 01 (Hokkaidō) |
|
| 36 |
+
| **region:KR** | Uppercase subdivision code complying to ISO 3166-2:KR. | 11 (Seoul), 26 (Busan), 41 (Gyeonggi) |
|
chat_template.jinja
ADDED
|
@@ -0,0 +1,139 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- if not messages %}
|
| 2 |
+
{{- raise_exception('No messages provided.') }}
|
| 3 |
+
{%- endif %}
|
| 4 |
+
{%- set image_count = namespace(value=0) %}
|
| 5 |
+
{%- set image_placeholder = '<|vision_start|><|image_pad|><|vision_end|>' -%}
|
| 6 |
+
{%- set mode = mode | default('content') -%}
|
| 7 |
+
{%- if template -%}{%- set mode = 'structured' -%}{%- endif -%}
|
| 8 |
+
{%- if not template and mode == 'structured' %}
|
| 9 |
+
{{- raise_exception('`structured` mode specified but no `template` provided.') }}
|
| 10 |
+
{%- endif %}
|
| 11 |
+
{%- if mode not in ['structured', 'content', 'template-generation', 'document-detection', 'markdown'] -%}{%- set mode = 'content' -%}{%- endif -%}
|
| 12 |
+
{%- if mode == 'markdown' %}{%- set mode = 'content' -%}{%- endif %}
|
| 13 |
+
{%- set enable_thinking = enable_thinking | default(False) -%}
|
| 14 |
+
{%- if mode not in ['structured', 'content'] and enable_thinking %}
|
| 15 |
+
{{- raise_exception('`enable_thinking` can only be `True` for `structured` and `content` modes.') }}
|
| 16 |
+
{%- endif %}
|
| 17 |
+
{%- set has_examples = namespace(flag=false) -%}
|
| 18 |
+
{%- if mode != 'structured' -%}{%- set has_examples = false -%}{%- endif -%}
|
| 19 |
+
{# MACRO TO RENDER MESSAGE CONTENT #}
|
| 20 |
+
{%- macro render_content(content, do_vision_count, is_system_content=false) %}
|
| 21 |
+
{%- if content is string %}
|
| 22 |
+
{{- content }}
|
| 23 |
+
{%- elif content is iterable and content is not mapping %}
|
| 24 |
+
{%- for item in content %}
|
| 25 |
+
{%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
|
| 26 |
+
{%- if is_system_content %}
|
| 27 |
+
{{- raise_exception('System message cannot contain images.') }}
|
| 28 |
+
{%- endif %}
|
| 29 |
+
{%- if do_vision_count %}
|
| 30 |
+
{%- set image_count.value = image_count.value + 1 %}
|
| 31 |
+
{%- endif %}
|
| 32 |
+
{%- if add_vision_id %}
|
| 33 |
+
{{- 'Picture ' ~ image_count.value ~ ': ' }}
|
| 34 |
+
{%- endif %}
|
| 35 |
+
{{- '<|vision_start|><|image_pad|><|vision_end|>\n' }}
|
| 36 |
+
{%- elif 'text' in item %}
|
| 37 |
+
{{- item.text + '\n' }}
|
| 38 |
+
{%- else %}
|
| 39 |
+
{{- raise_exception('Unexpected item type in content.') }}
|
| 40 |
+
{%- endif %}
|
| 41 |
+
{%- endfor %}
|
| 42 |
+
{%- elif content is none or content is undefined %}
|
| 43 |
+
{{- '' }}
|
| 44 |
+
{%- else %}
|
| 45 |
+
{{- raise_exception('Unexpected content type.') }}
|
| 46 |
+
{%- endif %}
|
| 47 |
+
{%- endmacro %}
|
| 48 |
+
{# SYSTEM MESSAGE #}
|
| 49 |
+
{%- if messages[0].role == 'system' %}
|
| 50 |
+
{%- set content = render_content(messages[0].content, false, true)|trim %}
|
| 51 |
+
{{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
|
| 52 |
+
{%- endif %}
|
| 53 |
+
{# USER MESSAGE #}
|
| 54 |
+
{{- '<|im_start|>user\n' -}}
|
| 55 |
+
{{- '【task】' + mode|replace("-", " ") + '\n' -}}
|
| 56 |
+
{# Template Section (for structured task): specifies template, instructions, examples, previous_output #}
|
| 57 |
+
{%- if mode == 'structured' -%}
|
| 58 |
+
{{- '【template_start】' + template + '【template_end】\n' -}}
|
| 59 |
+
{# Instructions Section #}
|
| 60 |
+
{%- if instructions -%}
|
| 61 |
+
{{- '【instructions_start】' + instructions + '【instructions_end】\n'-}}
|
| 62 |
+
{%- endif -%}
|
| 63 |
+
{# Examples Section (only for extraction tasks) #}
|
| 64 |
+
{%- for message in messages -%}
|
| 65 |
+
{%- if message.role == 'developer' and 'content' in message -%}
|
| 66 |
+
{# Validate that there is at least one input and one output contents #}
|
| 67 |
+
{%- set example_inputs = message.content[:-1] -%}
|
| 68 |
+
{%- set example_output_part = message.content[-1] -%}
|
| 69 |
+
{%- if example_inputs|length > 0 -%}
|
| 70 |
+
{%- if not has_examples.flag -%}
|
| 71 |
+
{{- '【examples_start】\n' -}}
|
| 72 |
+
{%- set has_examples.flag = true -%}
|
| 73 |
+
{%- endif -%}
|
| 74 |
+
{{- '【example_input_start】' + render_content(example_inputs, true)|trim + '【example_input_end】\n' -}}
|
| 75 |
+
{# Example output: only keep the text of the first output content #}
|
| 76 |
+
{%- set output_text = '' -%}
|
| 77 |
+
{%- if example_output_part is string -%}
|
| 78 |
+
{%- set output_text = example_output_part -%}
|
| 79 |
+
{%- elif example_output_part.text is defined -%}
|
| 80 |
+
{%- set output_text = example_output_part.text -%}
|
| 81 |
+
{%- endif -%}
|
| 82 |
+
{{- '【example_output_start】' + output_text|trim + '【example_output_end】\n' -}}
|
| 83 |
+
{%- if loop.last and has_examples.flag -%}
|
| 84 |
+
{{- '【examples_end】\n' -}}
|
| 85 |
+
{%- endif -%}
|
| 86 |
+
{%- endif -%}
|
| 87 |
+
{%- endif -%}
|
| 88 |
+
{%- endfor -%}
|
| 89 |
+
{# Previous Output Section #}
|
| 90 |
+
{%- if previous_output -%}
|
| 91 |
+
{{- '【previous_output_start】' + previous_output + '【previous_output_end】\n' -}}
|
| 92 |
+
{%- endif -%}
|
| 93 |
+
{%- endif -%}
|
| 94 |
+
{{- '【document_start】\n' -}}
|
| 95 |
+
{# PROCESS PROVIDED USER MESSAGES (RENDERED INTO A SINGLE ONE) #}
|
| 96 |
+
{%- for message in messages -%}
|
| 97 |
+
{%- if message.role == "system" %}
|
| 98 |
+
{%- if not loop.first %}
|
| 99 |
+
{{- raise_exception('System message must be at the beginning.') }}
|
| 100 |
+
{%- endif %}
|
| 101 |
+
{%- elif message.role == 'user' and message.name != "example" -%}
|
| 102 |
+
{%- set content = render_content(message.content, true)|trim %}
|
| 103 |
+
{{- content + '\n' -}}
|
| 104 |
+
{# {%- elif message.role == 'assistant' and not loop.last %}
|
| 105 |
+
llama.cpp renders a synthetic init example with an assistant turn in
|
| 106 |
+
the middle; ignore it so valid NuExtract prompts render unchanged.
|
| 107 |
+
{{- raise_exception('Assistant message must be at the end.') }} #}
|
| 108 |
+
{%- endif %}
|
| 109 |
+
{%- endfor -%}
|
| 110 |
+
{{- '【document_end】<|im_end|>\n' -}}
|
| 111 |
+
{# ASSISTANT MESSAGE #}
|
| 112 |
+
{%- if messages[-1].role == 'assistant' %}
|
| 113 |
+
{%- if add_generation_prompt -%}
|
| 114 |
+
{{- raise_exception('`add_generation_prompt` can only be `True` when no assistant message is provided.') }}
|
| 115 |
+
{%- endif %}
|
| 116 |
+
{%- set content = render_content(messages[-1].content, true)|trim %}
|
| 117 |
+
{%- set reasoning_content = '' %}
|
| 118 |
+
{%- if messages[-1].reasoning_content is string %}
|
| 119 |
+
{%- set reasoning_content = messages[-1].reasoning_content %}
|
| 120 |
+
{%- else %}
|
| 121 |
+
{%- if '</think>' in content %}
|
| 122 |
+
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
| 123 |
+
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
| 124 |
+
{%- endif %}
|
| 125 |
+
{%- endif %}
|
| 126 |
+
{%- set reasoning_content = reasoning_content|trim %}
|
| 127 |
+
{% generation %}
|
| 128 |
+
{{- '<|im_start|>assistant\n<think>\n' + reasoning_content + '\n</think>\n\n' + content + '<|im_end|>\n' -}}
|
| 129 |
+
{% endgeneration %}
|
| 130 |
+
{%- endif -%}
|
| 131 |
+
{# GENERATION PROMPT #}
|
| 132 |
+
{%- if add_generation_prompt -%}
|
| 133 |
+
{{- '<|im_start|>assistant\n' -}}
|
| 134 |
+
{%- if not enable_thinking -%}
|
| 135 |
+
{{- '<think>\n\n</think>\n\n' -}}
|
| 136 |
+
{%- else %}
|
| 137 |
+
{{- '<think>\n' -}}
|
| 138 |
+
{%- endif %}
|
| 139 |
+
{%- endif -%}
|
config.json
ADDED
|
@@ -0,0 +1,112 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen3_5ForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"bos_token_id": null,
|
| 6 |
+
"dtype": "bfloat16",
|
| 7 |
+
"eos_token_id": 248046,
|
| 8 |
+
"image_token_id": 248056,
|
| 9 |
+
"model_type": "qwen3_5",
|
| 10 |
+
"pad_token_id": 248044,
|
| 11 |
+
"text_config": {
|
| 12 |
+
"attention_bias": false,
|
| 13 |
+
"attention_dropout": 0.0,
|
| 14 |
+
"attn_output_gate": true,
|
| 15 |
+
"bos_token_id": null,
|
| 16 |
+
"dtype": "bfloat16",
|
| 17 |
+
"eos_token_id": 248044,
|
| 18 |
+
"full_attention_interval": 4,
|
| 19 |
+
"head_dim": 256,
|
| 20 |
+
"hidden_act": "silu",
|
| 21 |
+
"hidden_size": 2560,
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"intermediate_size": 9216,
|
| 24 |
+
"layer_types": [
|
| 25 |
+
"linear_attention",
|
| 26 |
+
"linear_attention",
|
| 27 |
+
"linear_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"linear_attention",
|
| 30 |
+
"linear_attention",
|
| 31 |
+
"linear_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"linear_attention",
|
| 34 |
+
"linear_attention",
|
| 35 |
+
"linear_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"linear_attention",
|
| 38 |
+
"linear_attention",
|
| 39 |
+
"linear_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"linear_attention",
|
| 42 |
+
"linear_attention",
|
| 43 |
+
"linear_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"linear_attention",
|
| 46 |
+
"linear_attention",
|
| 47 |
+
"linear_attention",
|
| 48 |
+
"full_attention",
|
| 49 |
+
"linear_attention",
|
| 50 |
+
"linear_attention",
|
| 51 |
+
"linear_attention",
|
| 52 |
+
"full_attention",
|
| 53 |
+
"linear_attention",
|
| 54 |
+
"linear_attention",
|
| 55 |
+
"linear_attention",
|
| 56 |
+
"full_attention"
|
| 57 |
+
],
|
| 58 |
+
"linear_conv_kernel_dim": 4,
|
| 59 |
+
"linear_key_head_dim": 128,
|
| 60 |
+
"linear_num_key_heads": 16,
|
| 61 |
+
"linear_num_value_heads": 32,
|
| 62 |
+
"linear_value_head_dim": 128,
|
| 63 |
+
"mamba_ssm_dtype": "float32",
|
| 64 |
+
"max_position_embeddings": 262144,
|
| 65 |
+
"mlp_only_layers": [],
|
| 66 |
+
"model_type": "qwen3_5_text",
|
| 67 |
+
"mtp_num_hidden_layers": 1,
|
| 68 |
+
"mtp_use_dedicated_embeddings": false,
|
| 69 |
+
"num_attention_heads": 16,
|
| 70 |
+
"num_hidden_layers": 32,
|
| 71 |
+
"num_key_value_heads": 4,
|
| 72 |
+
"pad_token_id": null,
|
| 73 |
+
"partial_rotary_factor": 0.25,
|
| 74 |
+
"rms_norm_eps": 1e-06,
|
| 75 |
+
"rope_parameters": {
|
| 76 |
+
"mrope_interleaved": true,
|
| 77 |
+
"mrope_section": [
|
| 78 |
+
11,
|
| 79 |
+
11,
|
| 80 |
+
10
|
| 81 |
+
],
|
| 82 |
+
"partial_rotary_factor": 0.25,
|
| 83 |
+
"rope_theta": 10000000,
|
| 84 |
+
"rope_type": "default"
|
| 85 |
+
},
|
| 86 |
+
"tie_word_embeddings": true,
|
| 87 |
+
"use_cache": true,
|
| 88 |
+
"vocab_size": 248320
|
| 89 |
+
},
|
| 90 |
+
"tie_word_embeddings": true,
|
| 91 |
+
"transformers_version": "5.5.4",
|
| 92 |
+
"video_token_id": 248057,
|
| 93 |
+
"vision_config": {
|
| 94 |
+
"deepstack_visual_indexes": [],
|
| 95 |
+
"depth": 24,
|
| 96 |
+
"dtype": "bfloat16",
|
| 97 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 98 |
+
"hidden_size": 1024,
|
| 99 |
+
"in_channels": 3,
|
| 100 |
+
"initializer_range": 0.02,
|
| 101 |
+
"intermediate_size": 4096,
|
| 102 |
+
"model_type": "qwen3_5",
|
| 103 |
+
"num_heads": 16,
|
| 104 |
+
"num_position_embeddings": 2304,
|
| 105 |
+
"out_hidden_size": 2560,
|
| 106 |
+
"patch_size": 16,
|
| 107 |
+
"spatial_merge_size": 2,
|
| 108 |
+
"temporal_patch_size": 2
|
| 109 |
+
},
|
| 110 |
+
"vision_end_token_id": 248054,
|
| 111 |
+
"vision_start_token_id": 248053
|
| 112 |
+
}
|
extraction_performance_no_thinking.png
ADDED
|
Git LFS Details
|
extraction_performance_thinking.png
ADDED
|
Git LFS Details
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"eos_token_id": 248044,
|
| 4 |
+
"transformers_version": "5.5.4",
|
| 5 |
+
"use_cache": true
|
| 6 |
+
}
|
header.svg
ADDED
|
logo.png
ADDED
|
Git LFS Details
|
logo_nuextract.svg
ADDED
|
markdown.png
ADDED
|
Git LFS Details
|
markdown_base.png
ADDED
|
Git LFS Details
|
markdown_reasoning.png
ADDED
|
Git LFS Details
|
md2st.svg
ADDED
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aca0a9d61da5df4fa4b1475b68c0a7205e5f8f5f20beb5055fde0622991f9ed7
|
| 3 |
+
size 9078620504
|
model.safetensors.index.json
ADDED
|
@@ -0,0 +1,745 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_size": 9319821208
|
| 4 |
+
},
|
| 5 |
+
"weight_map": {
|
| 6 |
+
"model.language_model.embed_tokens.weight": "model.safetensors",
|
| 7 |
+
"model.language_model.layers.0.input_layernorm.weight": "model.safetensors",
|
| 8 |
+
"model.language_model.layers.0.linear_attn.A_log": "model.safetensors",
|
| 9 |
+
"model.language_model.layers.0.linear_attn.conv1d.weight": "model.safetensors",
|
| 10 |
+
"model.language_model.layers.0.linear_attn.dt_bias": "model.safetensors",
|
| 11 |
+
"model.language_model.layers.0.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 12 |
+
"model.language_model.layers.0.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 13 |
+
"model.language_model.layers.0.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 14 |
+
"model.language_model.layers.0.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 15 |
+
"model.language_model.layers.0.linear_attn.norm.weight": "model.safetensors",
|
| 16 |
+
"model.language_model.layers.0.linear_attn.out_proj.weight": "model.safetensors",
|
| 17 |
+
"model.language_model.layers.0.mlp.down_proj.weight": "model.safetensors",
|
| 18 |
+
"model.language_model.layers.0.mlp.gate_proj.weight": "model.safetensors",
|
| 19 |
+
"model.language_model.layers.0.mlp.up_proj.weight": "model.safetensors",
|
| 20 |
+
"model.language_model.layers.0.post_attention_layernorm.weight": "model.safetensors",
|
| 21 |
+
"model.language_model.layers.1.input_layernorm.weight": "model.safetensors",
|
| 22 |
+
"model.language_model.layers.1.linear_attn.A_log": "model.safetensors",
|
| 23 |
+
"model.language_model.layers.1.linear_attn.conv1d.weight": "model.safetensors",
|
| 24 |
+
"model.language_model.layers.1.linear_attn.dt_bias": "model.safetensors",
|
| 25 |
+
"model.language_model.layers.1.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 26 |
+
"model.language_model.layers.1.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 27 |
+
"model.language_model.layers.1.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 28 |
+
"model.language_model.layers.1.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 29 |
+
"model.language_model.layers.1.linear_attn.norm.weight": "model.safetensors",
|
| 30 |
+
"model.language_model.layers.1.linear_attn.out_proj.weight": "model.safetensors",
|
| 31 |
+
"model.language_model.layers.1.mlp.down_proj.weight": "model.safetensors",
|
| 32 |
+
"model.language_model.layers.1.mlp.gate_proj.weight": "model.safetensors",
|
| 33 |
+
"model.language_model.layers.1.mlp.up_proj.weight": "model.safetensors",
|
| 34 |
+
"model.language_model.layers.1.post_attention_layernorm.weight": "model.safetensors",
|
| 35 |
+
"model.language_model.layers.10.input_layernorm.weight": "model.safetensors",
|
| 36 |
+
"model.language_model.layers.10.linear_attn.A_log": "model.safetensors",
|
| 37 |
+
"model.language_model.layers.10.linear_attn.conv1d.weight": "model.safetensors",
|
| 38 |
+
"model.language_model.layers.10.linear_attn.dt_bias": "model.safetensors",
|
| 39 |
+
"model.language_model.layers.10.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 40 |
+
"model.language_model.layers.10.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 41 |
+
"model.language_model.layers.10.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 42 |
+
"model.language_model.layers.10.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 43 |
+
"model.language_model.layers.10.linear_attn.norm.weight": "model.safetensors",
|
| 44 |
+
"model.language_model.layers.10.linear_attn.out_proj.weight": "model.safetensors",
|
| 45 |
+
"model.language_model.layers.10.mlp.down_proj.weight": "model.safetensors",
|
| 46 |
+
"model.language_model.layers.10.mlp.gate_proj.weight": "model.safetensors",
|
| 47 |
+
"model.language_model.layers.10.mlp.up_proj.weight": "model.safetensors",
|
| 48 |
+
"model.language_model.layers.10.post_attention_layernorm.weight": "model.safetensors",
|
| 49 |
+
"model.language_model.layers.11.input_layernorm.weight": "model.safetensors",
|
| 50 |
+
"model.language_model.layers.11.mlp.down_proj.weight": "model.safetensors",
|
| 51 |
+
"model.language_model.layers.11.mlp.gate_proj.weight": "model.safetensors",
|
| 52 |
+
"model.language_model.layers.11.mlp.up_proj.weight": "model.safetensors",
|
| 53 |
+
"model.language_model.layers.11.post_attention_layernorm.weight": "model.safetensors",
|
| 54 |
+
"model.language_model.layers.11.self_attn.k_norm.weight": "model.safetensors",
|
| 55 |
+
"model.language_model.layers.11.self_attn.k_proj.weight": "model.safetensors",
|
| 56 |
+
"model.language_model.layers.11.self_attn.o_proj.weight": "model.safetensors",
|
| 57 |
+
"model.language_model.layers.11.self_attn.q_norm.weight": "model.safetensors",
|
| 58 |
+
"model.language_model.layers.11.self_attn.q_proj.weight": "model.safetensors",
|
| 59 |
+
"model.language_model.layers.11.self_attn.v_proj.weight": "model.safetensors",
|
| 60 |
+
"model.language_model.layers.12.input_layernorm.weight": "model.safetensors",
|
| 61 |
+
"model.language_model.layers.12.linear_attn.A_log": "model.safetensors",
|
| 62 |
+
"model.language_model.layers.12.linear_attn.conv1d.weight": "model.safetensors",
|
| 63 |
+
"model.language_model.layers.12.linear_attn.dt_bias": "model.safetensors",
|
| 64 |
+
"model.language_model.layers.12.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 65 |
+
"model.language_model.layers.12.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 66 |
+
"model.language_model.layers.12.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 67 |
+
"model.language_model.layers.12.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 68 |
+
"model.language_model.layers.12.linear_attn.norm.weight": "model.safetensors",
|
| 69 |
+
"model.language_model.layers.12.linear_attn.out_proj.weight": "model.safetensors",
|
| 70 |
+
"model.language_model.layers.12.mlp.down_proj.weight": "model.safetensors",
|
| 71 |
+
"model.language_model.layers.12.mlp.gate_proj.weight": "model.safetensors",
|
| 72 |
+
"model.language_model.layers.12.mlp.up_proj.weight": "model.safetensors",
|
| 73 |
+
"model.language_model.layers.12.post_attention_layernorm.weight": "model.safetensors",
|
| 74 |
+
"model.language_model.layers.13.input_layernorm.weight": "model.safetensors",
|
| 75 |
+
"model.language_model.layers.13.linear_attn.A_log": "model.safetensors",
|
| 76 |
+
"model.language_model.layers.13.linear_attn.conv1d.weight": "model.safetensors",
|
| 77 |
+
"model.language_model.layers.13.linear_attn.dt_bias": "model.safetensors",
|
| 78 |
+
"model.language_model.layers.13.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 79 |
+
"model.language_model.layers.13.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 80 |
+
"model.language_model.layers.13.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 81 |
+
"model.language_model.layers.13.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 82 |
+
"model.language_model.layers.13.linear_attn.norm.weight": "model.safetensors",
|
| 83 |
+
"model.language_model.layers.13.linear_attn.out_proj.weight": "model.safetensors",
|
| 84 |
+
"model.language_model.layers.13.mlp.down_proj.weight": "model.safetensors",
|
| 85 |
+
"model.language_model.layers.13.mlp.gate_proj.weight": "model.safetensors",
|
| 86 |
+
"model.language_model.layers.13.mlp.up_proj.weight": "model.safetensors",
|
| 87 |
+
"model.language_model.layers.13.post_attention_layernorm.weight": "model.safetensors",
|
| 88 |
+
"model.language_model.layers.14.input_layernorm.weight": "model.safetensors",
|
| 89 |
+
"model.language_model.layers.14.linear_attn.A_log": "model.safetensors",
|
| 90 |
+
"model.language_model.layers.14.linear_attn.conv1d.weight": "model.safetensors",
|
| 91 |
+
"model.language_model.layers.14.linear_attn.dt_bias": "model.safetensors",
|
| 92 |
+
"model.language_model.layers.14.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 93 |
+
"model.language_model.layers.14.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 94 |
+
"model.language_model.layers.14.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 95 |
+
"model.language_model.layers.14.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 96 |
+
"model.language_model.layers.14.linear_attn.norm.weight": "model.safetensors",
|
| 97 |
+
"model.language_model.layers.14.linear_attn.out_proj.weight": "model.safetensors",
|
| 98 |
+
"model.language_model.layers.14.mlp.down_proj.weight": "model.safetensors",
|
| 99 |
+
"model.language_model.layers.14.mlp.gate_proj.weight": "model.safetensors",
|
| 100 |
+
"model.language_model.layers.14.mlp.up_proj.weight": "model.safetensors",
|
| 101 |
+
"model.language_model.layers.14.post_attention_layernorm.weight": "model.safetensors",
|
| 102 |
+
"model.language_model.layers.15.input_layernorm.weight": "model.safetensors",
|
| 103 |
+
"model.language_model.layers.15.mlp.down_proj.weight": "model.safetensors",
|
| 104 |
+
"model.language_model.layers.15.mlp.gate_proj.weight": "model.safetensors",
|
| 105 |
+
"model.language_model.layers.15.mlp.up_proj.weight": "model.safetensors",
|
| 106 |
+
"model.language_model.layers.15.post_attention_layernorm.weight": "model.safetensors",
|
| 107 |
+
"model.language_model.layers.15.self_attn.k_norm.weight": "model.safetensors",
|
| 108 |
+
"model.language_model.layers.15.self_attn.k_proj.weight": "model.safetensors",
|
| 109 |
+
"model.language_model.layers.15.self_attn.o_proj.weight": "model.safetensors",
|
| 110 |
+
"model.language_model.layers.15.self_attn.q_norm.weight": "model.safetensors",
|
| 111 |
+
"model.language_model.layers.15.self_attn.q_proj.weight": "model.safetensors",
|
| 112 |
+
"model.language_model.layers.15.self_attn.v_proj.weight": "model.safetensors",
|
| 113 |
+
"model.language_model.layers.16.input_layernorm.weight": "model.safetensors",
|
| 114 |
+
"model.language_model.layers.16.linear_attn.A_log": "model.safetensors",
|
| 115 |
+
"model.language_model.layers.16.linear_attn.conv1d.weight": "model.safetensors",
|
| 116 |
+
"model.language_model.layers.16.linear_attn.dt_bias": "model.safetensors",
|
| 117 |
+
"model.language_model.layers.16.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 118 |
+
"model.language_model.layers.16.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 119 |
+
"model.language_model.layers.16.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 120 |
+
"model.language_model.layers.16.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 121 |
+
"model.language_model.layers.16.linear_attn.norm.weight": "model.safetensors",
|
| 122 |
+
"model.language_model.layers.16.linear_attn.out_proj.weight": "model.safetensors",
|
| 123 |
+
"model.language_model.layers.16.mlp.down_proj.weight": "model.safetensors",
|
| 124 |
+
"model.language_model.layers.16.mlp.gate_proj.weight": "model.safetensors",
|
| 125 |
+
"model.language_model.layers.16.mlp.up_proj.weight": "model.safetensors",
|
| 126 |
+
"model.language_model.layers.16.post_attention_layernorm.weight": "model.safetensors",
|
| 127 |
+
"model.language_model.layers.17.input_layernorm.weight": "model.safetensors",
|
| 128 |
+
"model.language_model.layers.17.linear_attn.A_log": "model.safetensors",
|
| 129 |
+
"model.language_model.layers.17.linear_attn.conv1d.weight": "model.safetensors",
|
| 130 |
+
"model.language_model.layers.17.linear_attn.dt_bias": "model.safetensors",
|
| 131 |
+
"model.language_model.layers.17.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 132 |
+
"model.language_model.layers.17.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 133 |
+
"model.language_model.layers.17.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 134 |
+
"model.language_model.layers.17.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 135 |
+
"model.language_model.layers.17.linear_attn.norm.weight": "model.safetensors",
|
| 136 |
+
"model.language_model.layers.17.linear_attn.out_proj.weight": "model.safetensors",
|
| 137 |
+
"model.language_model.layers.17.mlp.down_proj.weight": "model.safetensors",
|
| 138 |
+
"model.language_model.layers.17.mlp.gate_proj.weight": "model.safetensors",
|
| 139 |
+
"model.language_model.layers.17.mlp.up_proj.weight": "model.safetensors",
|
| 140 |
+
"model.language_model.layers.17.post_attention_layernorm.weight": "model.safetensors",
|
| 141 |
+
"model.language_model.layers.18.input_layernorm.weight": "model.safetensors",
|
| 142 |
+
"model.language_model.layers.18.linear_attn.A_log": "model.safetensors",
|
| 143 |
+
"model.language_model.layers.18.linear_attn.conv1d.weight": "model.safetensors",
|
| 144 |
+
"model.language_model.layers.18.linear_attn.dt_bias": "model.safetensors",
|
| 145 |
+
"model.language_model.layers.18.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 146 |
+
"model.language_model.layers.18.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 147 |
+
"model.language_model.layers.18.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 148 |
+
"model.language_model.layers.18.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 149 |
+
"model.language_model.layers.18.linear_attn.norm.weight": "model.safetensors",
|
| 150 |
+
"model.language_model.layers.18.linear_attn.out_proj.weight": "model.safetensors",
|
| 151 |
+
"model.language_model.layers.18.mlp.down_proj.weight": "model.safetensors",
|
| 152 |
+
"model.language_model.layers.18.mlp.gate_proj.weight": "model.safetensors",
|
| 153 |
+
"model.language_model.layers.18.mlp.up_proj.weight": "model.safetensors",
|
| 154 |
+
"model.language_model.layers.18.post_attention_layernorm.weight": "model.safetensors",
|
| 155 |
+
"model.language_model.layers.19.input_layernorm.weight": "model.safetensors",
|
| 156 |
+
"model.language_model.layers.19.mlp.down_proj.weight": "model.safetensors",
|
| 157 |
+
"model.language_model.layers.19.mlp.gate_proj.weight": "model.safetensors",
|
| 158 |
+
"model.language_model.layers.19.mlp.up_proj.weight": "model.safetensors",
|
| 159 |
+
"model.language_model.layers.19.post_attention_layernorm.weight": "model.safetensors",
|
| 160 |
+
"model.language_model.layers.19.self_attn.k_norm.weight": "model.safetensors",
|
| 161 |
+
"model.language_model.layers.19.self_attn.k_proj.weight": "model.safetensors",
|
| 162 |
+
"model.language_model.layers.19.self_attn.o_proj.weight": "model.safetensors",
|
| 163 |
+
"model.language_model.layers.19.self_attn.q_norm.weight": "model.safetensors",
|
| 164 |
+
"model.language_model.layers.19.self_attn.q_proj.weight": "model.safetensors",
|
| 165 |
+
"model.language_model.layers.19.self_attn.v_proj.weight": "model.safetensors",
|
| 166 |
+
"model.language_model.layers.2.input_layernorm.weight": "model.safetensors",
|
| 167 |
+
"model.language_model.layers.2.linear_attn.A_log": "model.safetensors",
|
| 168 |
+
"model.language_model.layers.2.linear_attn.conv1d.weight": "model.safetensors",
|
| 169 |
+
"model.language_model.layers.2.linear_attn.dt_bias": "model.safetensors",
|
| 170 |
+
"model.language_model.layers.2.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 171 |
+
"model.language_model.layers.2.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 172 |
+
"model.language_model.layers.2.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 173 |
+
"model.language_model.layers.2.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 174 |
+
"model.language_model.layers.2.linear_attn.norm.weight": "model.safetensors",
|
| 175 |
+
"model.language_model.layers.2.linear_attn.out_proj.weight": "model.safetensors",
|
| 176 |
+
"model.language_model.layers.2.mlp.down_proj.weight": "model.safetensors",
|
| 177 |
+
"model.language_model.layers.2.mlp.gate_proj.weight": "model.safetensors",
|
| 178 |
+
"model.language_model.layers.2.mlp.up_proj.weight": "model.safetensors",
|
| 179 |
+
"model.language_model.layers.2.post_attention_layernorm.weight": "model.safetensors",
|
| 180 |
+
"model.language_model.layers.20.input_layernorm.weight": "model.safetensors",
|
| 181 |
+
"model.language_model.layers.20.linear_attn.A_log": "model.safetensors",
|
| 182 |
+
"model.language_model.layers.20.linear_attn.conv1d.weight": "model.safetensors",
|
| 183 |
+
"model.language_model.layers.20.linear_attn.dt_bias": "model.safetensors",
|
| 184 |
+
"model.language_model.layers.20.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 185 |
+
"model.language_model.layers.20.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 186 |
+
"model.language_model.layers.20.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 187 |
+
"model.language_model.layers.20.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 188 |
+
"model.language_model.layers.20.linear_attn.norm.weight": "model.safetensors",
|
| 189 |
+
"model.language_model.layers.20.linear_attn.out_proj.weight": "model.safetensors",
|
| 190 |
+
"model.language_model.layers.20.mlp.down_proj.weight": "model.safetensors",
|
| 191 |
+
"model.language_model.layers.20.mlp.gate_proj.weight": "model.safetensors",
|
| 192 |
+
"model.language_model.layers.20.mlp.up_proj.weight": "model.safetensors",
|
| 193 |
+
"model.language_model.layers.20.post_attention_layernorm.weight": "model.safetensors",
|
| 194 |
+
"model.language_model.layers.21.input_layernorm.weight": "model.safetensors",
|
| 195 |
+
"model.language_model.layers.21.linear_attn.A_log": "model.safetensors",
|
| 196 |
+
"model.language_model.layers.21.linear_attn.conv1d.weight": "model.safetensors",
|
| 197 |
+
"model.language_model.layers.21.linear_attn.dt_bias": "model.safetensors",
|
| 198 |
+
"model.language_model.layers.21.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 199 |
+
"model.language_model.layers.21.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 200 |
+
"model.language_model.layers.21.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 201 |
+
"model.language_model.layers.21.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 202 |
+
"model.language_model.layers.21.linear_attn.norm.weight": "model.safetensors",
|
| 203 |
+
"model.language_model.layers.21.linear_attn.out_proj.weight": "model.safetensors",
|
| 204 |
+
"model.language_model.layers.21.mlp.down_proj.weight": "model.safetensors",
|
| 205 |
+
"model.language_model.layers.21.mlp.gate_proj.weight": "model.safetensors",
|
| 206 |
+
"model.language_model.layers.21.mlp.up_proj.weight": "model.safetensors",
|
| 207 |
+
"model.language_model.layers.21.post_attention_layernorm.weight": "model.safetensors",
|
| 208 |
+
"model.language_model.layers.22.input_layernorm.weight": "model.safetensors",
|
| 209 |
+
"model.language_model.layers.22.linear_attn.A_log": "model.safetensors",
|
| 210 |
+
"model.language_model.layers.22.linear_attn.conv1d.weight": "model.safetensors",
|
| 211 |
+
"model.language_model.layers.22.linear_attn.dt_bias": "model.safetensors",
|
| 212 |
+
"model.language_model.layers.22.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 213 |
+
"model.language_model.layers.22.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 214 |
+
"model.language_model.layers.22.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 215 |
+
"model.language_model.layers.22.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 216 |
+
"model.language_model.layers.22.linear_attn.norm.weight": "model.safetensors",
|
| 217 |
+
"model.language_model.layers.22.linear_attn.out_proj.weight": "model.safetensors",
|
| 218 |
+
"model.language_model.layers.22.mlp.down_proj.weight": "model.safetensors",
|
| 219 |
+
"model.language_model.layers.22.mlp.gate_proj.weight": "model.safetensors",
|
| 220 |
+
"model.language_model.layers.22.mlp.up_proj.weight": "model.safetensors",
|
| 221 |
+
"model.language_model.layers.22.post_attention_layernorm.weight": "model.safetensors",
|
| 222 |
+
"model.language_model.layers.23.input_layernorm.weight": "model.safetensors",
|
| 223 |
+
"model.language_model.layers.23.mlp.down_proj.weight": "model.safetensors",
|
| 224 |
+
"model.language_model.layers.23.mlp.gate_proj.weight": "model.safetensors",
|
| 225 |
+
"model.language_model.layers.23.mlp.up_proj.weight": "model.safetensors",
|
| 226 |
+
"model.language_model.layers.23.post_attention_layernorm.weight": "model.safetensors",
|
| 227 |
+
"model.language_model.layers.23.self_attn.k_norm.weight": "model.safetensors",
|
| 228 |
+
"model.language_model.layers.23.self_attn.k_proj.weight": "model.safetensors",
|
| 229 |
+
"model.language_model.layers.23.self_attn.o_proj.weight": "model.safetensors",
|
| 230 |
+
"model.language_model.layers.23.self_attn.q_norm.weight": "model.safetensors",
|
| 231 |
+
"model.language_model.layers.23.self_attn.q_proj.weight": "model.safetensors",
|
| 232 |
+
"model.language_model.layers.23.self_attn.v_proj.weight": "model.safetensors",
|
| 233 |
+
"model.language_model.layers.24.input_layernorm.weight": "model.safetensors",
|
| 234 |
+
"model.language_model.layers.24.linear_attn.A_log": "model.safetensors",
|
| 235 |
+
"model.language_model.layers.24.linear_attn.conv1d.weight": "model.safetensors",
|
| 236 |
+
"model.language_model.layers.24.linear_attn.dt_bias": "model.safetensors",
|
| 237 |
+
"model.language_model.layers.24.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 238 |
+
"model.language_model.layers.24.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 239 |
+
"model.language_model.layers.24.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 240 |
+
"model.language_model.layers.24.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 241 |
+
"model.language_model.layers.24.linear_attn.norm.weight": "model.safetensors",
|
| 242 |
+
"model.language_model.layers.24.linear_attn.out_proj.weight": "model.safetensors",
|
| 243 |
+
"model.language_model.layers.24.mlp.down_proj.weight": "model.safetensors",
|
| 244 |
+
"model.language_model.layers.24.mlp.gate_proj.weight": "model.safetensors",
|
| 245 |
+
"model.language_model.layers.24.mlp.up_proj.weight": "model.safetensors",
|
| 246 |
+
"model.language_model.layers.24.post_attention_layernorm.weight": "model.safetensors",
|
| 247 |
+
"model.language_model.layers.25.input_layernorm.weight": "model.safetensors",
|
| 248 |
+
"model.language_model.layers.25.linear_attn.A_log": "model.safetensors",
|
| 249 |
+
"model.language_model.layers.25.linear_attn.conv1d.weight": "model.safetensors",
|
| 250 |
+
"model.language_model.layers.25.linear_attn.dt_bias": "model.safetensors",
|
| 251 |
+
"model.language_model.layers.25.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 252 |
+
"model.language_model.layers.25.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 253 |
+
"model.language_model.layers.25.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 254 |
+
"model.language_model.layers.25.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 255 |
+
"model.language_model.layers.25.linear_attn.norm.weight": "model.safetensors",
|
| 256 |
+
"model.language_model.layers.25.linear_attn.out_proj.weight": "model.safetensors",
|
| 257 |
+
"model.language_model.layers.25.mlp.down_proj.weight": "model.safetensors",
|
| 258 |
+
"model.language_model.layers.25.mlp.gate_proj.weight": "model.safetensors",
|
| 259 |
+
"model.language_model.layers.25.mlp.up_proj.weight": "model.safetensors",
|
| 260 |
+
"model.language_model.layers.25.post_attention_layernorm.weight": "model.safetensors",
|
| 261 |
+
"model.language_model.layers.26.input_layernorm.weight": "model.safetensors",
|
| 262 |
+
"model.language_model.layers.26.linear_attn.A_log": "model.safetensors",
|
| 263 |
+
"model.language_model.layers.26.linear_attn.conv1d.weight": "model.safetensors",
|
| 264 |
+
"model.language_model.layers.26.linear_attn.dt_bias": "model.safetensors",
|
| 265 |
+
"model.language_model.layers.26.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 266 |
+
"model.language_model.layers.26.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 267 |
+
"model.language_model.layers.26.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 268 |
+
"model.language_model.layers.26.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 269 |
+
"model.language_model.layers.26.linear_attn.norm.weight": "model.safetensors",
|
| 270 |
+
"model.language_model.layers.26.linear_attn.out_proj.weight": "model.safetensors",
|
| 271 |
+
"model.language_model.layers.26.mlp.down_proj.weight": "model.safetensors",
|
| 272 |
+
"model.language_model.layers.26.mlp.gate_proj.weight": "model.safetensors",
|
| 273 |
+
"model.language_model.layers.26.mlp.up_proj.weight": "model.safetensors",
|
| 274 |
+
"model.language_model.layers.26.post_attention_layernorm.weight": "model.safetensors",
|
| 275 |
+
"model.language_model.layers.27.input_layernorm.weight": "model.safetensors",
|
| 276 |
+
"model.language_model.layers.27.mlp.down_proj.weight": "model.safetensors",
|
| 277 |
+
"model.language_model.layers.27.mlp.gate_proj.weight": "model.safetensors",
|
| 278 |
+
"model.language_model.layers.27.mlp.up_proj.weight": "model.safetensors",
|
| 279 |
+
"model.language_model.layers.27.post_attention_layernorm.weight": "model.safetensors",
|
| 280 |
+
"model.language_model.layers.27.self_attn.k_norm.weight": "model.safetensors",
|
| 281 |
+
"model.language_model.layers.27.self_attn.k_proj.weight": "model.safetensors",
|
| 282 |
+
"model.language_model.layers.27.self_attn.o_proj.weight": "model.safetensors",
|
| 283 |
+
"model.language_model.layers.27.self_attn.q_norm.weight": "model.safetensors",
|
| 284 |
+
"model.language_model.layers.27.self_attn.q_proj.weight": "model.safetensors",
|
| 285 |
+
"model.language_model.layers.27.self_attn.v_proj.weight": "model.safetensors",
|
| 286 |
+
"model.language_model.layers.28.input_layernorm.weight": "model.safetensors",
|
| 287 |
+
"model.language_model.layers.28.linear_attn.A_log": "model.safetensors",
|
| 288 |
+
"model.language_model.layers.28.linear_attn.conv1d.weight": "model.safetensors",
|
| 289 |
+
"model.language_model.layers.28.linear_attn.dt_bias": "model.safetensors",
|
| 290 |
+
"model.language_model.layers.28.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 291 |
+
"model.language_model.layers.28.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 292 |
+
"model.language_model.layers.28.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 293 |
+
"model.language_model.layers.28.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 294 |
+
"model.language_model.layers.28.linear_attn.norm.weight": "model.safetensors",
|
| 295 |
+
"model.language_model.layers.28.linear_attn.out_proj.weight": "model.safetensors",
|
| 296 |
+
"model.language_model.layers.28.mlp.down_proj.weight": "model.safetensors",
|
| 297 |
+
"model.language_model.layers.28.mlp.gate_proj.weight": "model.safetensors",
|
| 298 |
+
"model.language_model.layers.28.mlp.up_proj.weight": "model.safetensors",
|
| 299 |
+
"model.language_model.layers.28.post_attention_layernorm.weight": "model.safetensors",
|
| 300 |
+
"model.language_model.layers.29.input_layernorm.weight": "model.safetensors",
|
| 301 |
+
"model.language_model.layers.29.linear_attn.A_log": "model.safetensors",
|
| 302 |
+
"model.language_model.layers.29.linear_attn.conv1d.weight": "model.safetensors",
|
| 303 |
+
"model.language_model.layers.29.linear_attn.dt_bias": "model.safetensors",
|
| 304 |
+
"model.language_model.layers.29.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 305 |
+
"model.language_model.layers.29.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 306 |
+
"model.language_model.layers.29.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 307 |
+
"model.language_model.layers.29.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 308 |
+
"model.language_model.layers.29.linear_attn.norm.weight": "model.safetensors",
|
| 309 |
+
"model.language_model.layers.29.linear_attn.out_proj.weight": "model.safetensors",
|
| 310 |
+
"model.language_model.layers.29.mlp.down_proj.weight": "model.safetensors",
|
| 311 |
+
"model.language_model.layers.29.mlp.gate_proj.weight": "model.safetensors",
|
| 312 |
+
"model.language_model.layers.29.mlp.up_proj.weight": "model.safetensors",
|
| 313 |
+
"model.language_model.layers.29.post_attention_layernorm.weight": "model.safetensors",
|
| 314 |
+
"model.language_model.layers.3.input_layernorm.weight": "model.safetensors",
|
| 315 |
+
"model.language_model.layers.3.mlp.down_proj.weight": "model.safetensors",
|
| 316 |
+
"model.language_model.layers.3.mlp.gate_proj.weight": "model.safetensors",
|
| 317 |
+
"model.language_model.layers.3.mlp.up_proj.weight": "model.safetensors",
|
| 318 |
+
"model.language_model.layers.3.post_attention_layernorm.weight": "model.safetensors",
|
| 319 |
+
"model.language_model.layers.3.self_attn.k_norm.weight": "model.safetensors",
|
| 320 |
+
"model.language_model.layers.3.self_attn.k_proj.weight": "model.safetensors",
|
| 321 |
+
"model.language_model.layers.3.self_attn.o_proj.weight": "model.safetensors",
|
| 322 |
+
"model.language_model.layers.3.self_attn.q_norm.weight": "model.safetensors",
|
| 323 |
+
"model.language_model.layers.3.self_attn.q_proj.weight": "model.safetensors",
|
| 324 |
+
"model.language_model.layers.3.self_attn.v_proj.weight": "model.safetensors",
|
| 325 |
+
"model.language_model.layers.30.input_layernorm.weight": "model.safetensors",
|
| 326 |
+
"model.language_model.layers.30.linear_attn.A_log": "model.safetensors",
|
| 327 |
+
"model.language_model.layers.30.linear_attn.conv1d.weight": "model.safetensors",
|
| 328 |
+
"model.language_model.layers.30.linear_attn.dt_bias": "model.safetensors",
|
| 329 |
+
"model.language_model.layers.30.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 330 |
+
"model.language_model.layers.30.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 331 |
+
"model.language_model.layers.30.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 332 |
+
"model.language_model.layers.30.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 333 |
+
"model.language_model.layers.30.linear_attn.norm.weight": "model.safetensors",
|
| 334 |
+
"model.language_model.layers.30.linear_attn.out_proj.weight": "model.safetensors",
|
| 335 |
+
"model.language_model.layers.30.mlp.down_proj.weight": "model.safetensors",
|
| 336 |
+
"model.language_model.layers.30.mlp.gate_proj.weight": "model.safetensors",
|
| 337 |
+
"model.language_model.layers.30.mlp.up_proj.weight": "model.safetensors",
|
| 338 |
+
"model.language_model.layers.30.post_attention_layernorm.weight": "model.safetensors",
|
| 339 |
+
"model.language_model.layers.31.input_layernorm.weight": "model.safetensors",
|
| 340 |
+
"model.language_model.layers.31.mlp.down_proj.weight": "model.safetensors",
|
| 341 |
+
"model.language_model.layers.31.mlp.gate_proj.weight": "model.safetensors",
|
| 342 |
+
"model.language_model.layers.31.mlp.up_proj.weight": "model.safetensors",
|
| 343 |
+
"model.language_model.layers.31.post_attention_layernorm.weight": "model.safetensors",
|
| 344 |
+
"model.language_model.layers.31.self_attn.k_norm.weight": "model.safetensors",
|
| 345 |
+
"model.language_model.layers.31.self_attn.k_proj.weight": "model.safetensors",
|
| 346 |
+
"model.language_model.layers.31.self_attn.o_proj.weight": "model.safetensors",
|
| 347 |
+
"model.language_model.layers.31.self_attn.q_norm.weight": "model.safetensors",
|
| 348 |
+
"model.language_model.layers.31.self_attn.q_proj.weight": "model.safetensors",
|
| 349 |
+
"model.language_model.layers.31.self_attn.v_proj.weight": "model.safetensors",
|
| 350 |
+
"model.language_model.layers.4.input_layernorm.weight": "model.safetensors",
|
| 351 |
+
"model.language_model.layers.4.linear_attn.A_log": "model.safetensors",
|
| 352 |
+
"model.language_model.layers.4.linear_attn.conv1d.weight": "model.safetensors",
|
| 353 |
+
"model.language_model.layers.4.linear_attn.dt_bias": "model.safetensors",
|
| 354 |
+
"model.language_model.layers.4.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 355 |
+
"model.language_model.layers.4.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 356 |
+
"model.language_model.layers.4.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 357 |
+
"model.language_model.layers.4.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 358 |
+
"model.language_model.layers.4.linear_attn.norm.weight": "model.safetensors",
|
| 359 |
+
"model.language_model.layers.4.linear_attn.out_proj.weight": "model.safetensors",
|
| 360 |
+
"model.language_model.layers.4.mlp.down_proj.weight": "model.safetensors",
|
| 361 |
+
"model.language_model.layers.4.mlp.gate_proj.weight": "model.safetensors",
|
| 362 |
+
"model.language_model.layers.4.mlp.up_proj.weight": "model.safetensors",
|
| 363 |
+
"model.language_model.layers.4.post_attention_layernorm.weight": "model.safetensors",
|
| 364 |
+
"model.language_model.layers.5.input_layernorm.weight": "model.safetensors",
|
| 365 |
+
"model.language_model.layers.5.linear_attn.A_log": "model.safetensors",
|
| 366 |
+
"model.language_model.layers.5.linear_attn.conv1d.weight": "model.safetensors",
|
| 367 |
+
"model.language_model.layers.5.linear_attn.dt_bias": "model.safetensors",
|
| 368 |
+
"model.language_model.layers.5.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 369 |
+
"model.language_model.layers.5.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 370 |
+
"model.language_model.layers.5.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 371 |
+
"model.language_model.layers.5.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 372 |
+
"model.language_model.layers.5.linear_attn.norm.weight": "model.safetensors",
|
| 373 |
+
"model.language_model.layers.5.linear_attn.out_proj.weight": "model.safetensors",
|
| 374 |
+
"model.language_model.layers.5.mlp.down_proj.weight": "model.safetensors",
|
| 375 |
+
"model.language_model.layers.5.mlp.gate_proj.weight": "model.safetensors",
|
| 376 |
+
"model.language_model.layers.5.mlp.up_proj.weight": "model.safetensors",
|
| 377 |
+
"model.language_model.layers.5.post_attention_layernorm.weight": "model.safetensors",
|
| 378 |
+
"model.language_model.layers.6.input_layernorm.weight": "model.safetensors",
|
| 379 |
+
"model.language_model.layers.6.linear_attn.A_log": "model.safetensors",
|
| 380 |
+
"model.language_model.layers.6.linear_attn.conv1d.weight": "model.safetensors",
|
| 381 |
+
"model.language_model.layers.6.linear_attn.dt_bias": "model.safetensors",
|
| 382 |
+
"model.language_model.layers.6.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 383 |
+
"model.language_model.layers.6.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 384 |
+
"model.language_model.layers.6.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 385 |
+
"model.language_model.layers.6.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 386 |
+
"model.language_model.layers.6.linear_attn.norm.weight": "model.safetensors",
|
| 387 |
+
"model.language_model.layers.6.linear_attn.out_proj.weight": "model.safetensors",
|
| 388 |
+
"model.language_model.layers.6.mlp.down_proj.weight": "model.safetensors",
|
| 389 |
+
"model.language_model.layers.6.mlp.gate_proj.weight": "model.safetensors",
|
| 390 |
+
"model.language_model.layers.6.mlp.up_proj.weight": "model.safetensors",
|
| 391 |
+
"model.language_model.layers.6.post_attention_layernorm.weight": "model.safetensors",
|
| 392 |
+
"model.language_model.layers.7.input_layernorm.weight": "model.safetensors",
|
| 393 |
+
"model.language_model.layers.7.mlp.down_proj.weight": "model.safetensors",
|
| 394 |
+
"model.language_model.layers.7.mlp.gate_proj.weight": "model.safetensors",
|
| 395 |
+
"model.language_model.layers.7.mlp.up_proj.weight": "model.safetensors",
|
| 396 |
+
"model.language_model.layers.7.post_attention_layernorm.weight": "model.safetensors",
|
| 397 |
+
"model.language_model.layers.7.self_attn.k_norm.weight": "model.safetensors",
|
| 398 |
+
"model.language_model.layers.7.self_attn.k_proj.weight": "model.safetensors",
|
| 399 |
+
"model.language_model.layers.7.self_attn.o_proj.weight": "model.safetensors",
|
| 400 |
+
"model.language_model.layers.7.self_attn.q_norm.weight": "model.safetensors",
|
| 401 |
+
"model.language_model.layers.7.self_attn.q_proj.weight": "model.safetensors",
|
| 402 |
+
"model.language_model.layers.7.self_attn.v_proj.weight": "model.safetensors",
|
| 403 |
+
"model.language_model.layers.8.input_layernorm.weight": "model.safetensors",
|
| 404 |
+
"model.language_model.layers.8.linear_attn.A_log": "model.safetensors",
|
| 405 |
+
"model.language_model.layers.8.linear_attn.conv1d.weight": "model.safetensors",
|
| 406 |
+
"model.language_model.layers.8.linear_attn.dt_bias": "model.safetensors",
|
| 407 |
+
"model.language_model.layers.8.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 408 |
+
"model.language_model.layers.8.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 409 |
+
"model.language_model.layers.8.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 410 |
+
"model.language_model.layers.8.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 411 |
+
"model.language_model.layers.8.linear_attn.norm.weight": "model.safetensors",
|
| 412 |
+
"model.language_model.layers.8.linear_attn.out_proj.weight": "model.safetensors",
|
| 413 |
+
"model.language_model.layers.8.mlp.down_proj.weight": "model.safetensors",
|
| 414 |
+
"model.language_model.layers.8.mlp.gate_proj.weight": "model.safetensors",
|
| 415 |
+
"model.language_model.layers.8.mlp.up_proj.weight": "model.safetensors",
|
| 416 |
+
"model.language_model.layers.8.post_attention_layernorm.weight": "model.safetensors",
|
| 417 |
+
"model.language_model.layers.9.input_layernorm.weight": "model.safetensors",
|
| 418 |
+
"model.language_model.layers.9.linear_attn.A_log": "model.safetensors",
|
| 419 |
+
"model.language_model.layers.9.linear_attn.conv1d.weight": "model.safetensors",
|
| 420 |
+
"model.language_model.layers.9.linear_attn.dt_bias": "model.safetensors",
|
| 421 |
+
"model.language_model.layers.9.linear_attn.in_proj_a.weight": "model.safetensors",
|
| 422 |
+
"model.language_model.layers.9.linear_attn.in_proj_b.weight": "model.safetensors",
|
| 423 |
+
"model.language_model.layers.9.linear_attn.in_proj_qkv.weight": "model.safetensors",
|
| 424 |
+
"model.language_model.layers.9.linear_attn.in_proj_z.weight": "model.safetensors",
|
| 425 |
+
"model.language_model.layers.9.linear_attn.norm.weight": "model.safetensors",
|
| 426 |
+
"model.language_model.layers.9.linear_attn.out_proj.weight": "model.safetensors",
|
| 427 |
+
"model.language_model.layers.9.mlp.down_proj.weight": "model.safetensors",
|
| 428 |
+
"model.language_model.layers.9.mlp.gate_proj.weight": "model.safetensors",
|
| 429 |
+
"model.language_model.layers.9.mlp.up_proj.weight": "model.safetensors",
|
| 430 |
+
"model.language_model.layers.9.post_attention_layernorm.weight": "model.safetensors",
|
| 431 |
+
"model.language_model.norm.weight": "model.safetensors",
|
| 432 |
+
"model.visual.blocks.0.attn.proj.bias": "model.safetensors",
|
| 433 |
+
"model.visual.blocks.0.attn.proj.weight": "model.safetensors",
|
| 434 |
+
"model.visual.blocks.0.attn.qkv.bias": "model.safetensors",
|
| 435 |
+
"model.visual.blocks.0.attn.qkv.weight": "model.safetensors",
|
| 436 |
+
"model.visual.blocks.0.mlp.linear_fc1.bias": "model.safetensors",
|
| 437 |
+
"model.visual.blocks.0.mlp.linear_fc1.weight": "model.safetensors",
|
| 438 |
+
"model.visual.blocks.0.mlp.linear_fc2.bias": "model.safetensors",
|
| 439 |
+
"model.visual.blocks.0.mlp.linear_fc2.weight": "model.safetensors",
|
| 440 |
+
"model.visual.blocks.0.norm1.bias": "model.safetensors",
|
| 441 |
+
"model.visual.blocks.0.norm1.weight": "model.safetensors",
|
| 442 |
+
"model.visual.blocks.0.norm2.bias": "model.safetensors",
|
| 443 |
+
"model.visual.blocks.0.norm2.weight": "model.safetensors",
|
| 444 |
+
"model.visual.blocks.1.attn.proj.bias": "model.safetensors",
|
| 445 |
+
"model.visual.blocks.1.attn.proj.weight": "model.safetensors",
|
| 446 |
+
"model.visual.blocks.1.attn.qkv.bias": "model.safetensors",
|
| 447 |
+
"model.visual.blocks.1.attn.qkv.weight": "model.safetensors",
|
| 448 |
+
"model.visual.blocks.1.mlp.linear_fc1.bias": "model.safetensors",
|
| 449 |
+
"model.visual.blocks.1.mlp.linear_fc1.weight": "model.safetensors",
|
| 450 |
+
"model.visual.blocks.1.mlp.linear_fc2.bias": "model.safetensors",
|
| 451 |
+
"model.visual.blocks.1.mlp.linear_fc2.weight": "model.safetensors",
|
| 452 |
+
"model.visual.blocks.1.norm1.bias": "model.safetensors",
|
| 453 |
+
"model.visual.blocks.1.norm1.weight": "model.safetensors",
|
| 454 |
+
"model.visual.blocks.1.norm2.bias": "model.safetensors",
|
| 455 |
+
"model.visual.blocks.1.norm2.weight": "model.safetensors",
|
| 456 |
+
"model.visual.blocks.10.attn.proj.bias": "model.safetensors",
|
| 457 |
+
"model.visual.blocks.10.attn.proj.weight": "model.safetensors",
|
| 458 |
+
"model.visual.blocks.10.attn.qkv.bias": "model.safetensors",
|
| 459 |
+
"model.visual.blocks.10.attn.qkv.weight": "model.safetensors",
|
| 460 |
+
"model.visual.blocks.10.mlp.linear_fc1.bias": "model.safetensors",
|
| 461 |
+
"model.visual.blocks.10.mlp.linear_fc1.weight": "model.safetensors",
|
| 462 |
+
"model.visual.blocks.10.mlp.linear_fc2.bias": "model.safetensors",
|
| 463 |
+
"model.visual.blocks.10.mlp.linear_fc2.weight": "model.safetensors",
|
| 464 |
+
"model.visual.blocks.10.norm1.bias": "model.safetensors",
|
| 465 |
+
"model.visual.blocks.10.norm1.weight": "model.safetensors",
|
| 466 |
+
"model.visual.blocks.10.norm2.bias": "model.safetensors",
|
| 467 |
+
"model.visual.blocks.10.norm2.weight": "model.safetensors",
|
| 468 |
+
"model.visual.blocks.11.attn.proj.bias": "model.safetensors",
|
| 469 |
+
"model.visual.blocks.11.attn.proj.weight": "model.safetensors",
|
| 470 |
+
"model.visual.blocks.11.attn.qkv.bias": "model.safetensors",
|
| 471 |
+
"model.visual.blocks.11.attn.qkv.weight": "model.safetensors",
|
| 472 |
+
"model.visual.blocks.11.mlp.linear_fc1.bias": "model.safetensors",
|
| 473 |
+
"model.visual.blocks.11.mlp.linear_fc1.weight": "model.safetensors",
|
| 474 |
+
"model.visual.blocks.11.mlp.linear_fc2.bias": "model.safetensors",
|
| 475 |
+
"model.visual.blocks.11.mlp.linear_fc2.weight": "model.safetensors",
|
| 476 |
+
"model.visual.blocks.11.norm1.bias": "model.safetensors",
|
| 477 |
+
"model.visual.blocks.11.norm1.weight": "model.safetensors",
|
| 478 |
+
"model.visual.blocks.11.norm2.bias": "model.safetensors",
|
| 479 |
+
"model.visual.blocks.11.norm2.weight": "model.safetensors",
|
| 480 |
+
"model.visual.blocks.12.attn.proj.bias": "model.safetensors",
|
| 481 |
+
"model.visual.blocks.12.attn.proj.weight": "model.safetensors",
|
| 482 |
+
"model.visual.blocks.12.attn.qkv.bias": "model.safetensors",
|
| 483 |
+
"model.visual.blocks.12.attn.qkv.weight": "model.safetensors",
|
| 484 |
+
"model.visual.blocks.12.mlp.linear_fc1.bias": "model.safetensors",
|
| 485 |
+
"model.visual.blocks.12.mlp.linear_fc1.weight": "model.safetensors",
|
| 486 |
+
"model.visual.blocks.12.mlp.linear_fc2.bias": "model.safetensors",
|
| 487 |
+
"model.visual.blocks.12.mlp.linear_fc2.weight": "model.safetensors",
|
| 488 |
+
"model.visual.blocks.12.norm1.bias": "model.safetensors",
|
| 489 |
+
"model.visual.blocks.12.norm1.weight": "model.safetensors",
|
| 490 |
+
"model.visual.blocks.12.norm2.bias": "model.safetensors",
|
| 491 |
+
"model.visual.blocks.12.norm2.weight": "model.safetensors",
|
| 492 |
+
"model.visual.blocks.13.attn.proj.bias": "model.safetensors",
|
| 493 |
+
"model.visual.blocks.13.attn.proj.weight": "model.safetensors",
|
| 494 |
+
"model.visual.blocks.13.attn.qkv.bias": "model.safetensors",
|
| 495 |
+
"model.visual.blocks.13.attn.qkv.weight": "model.safetensors",
|
| 496 |
+
"model.visual.blocks.13.mlp.linear_fc1.bias": "model.safetensors",
|
| 497 |
+
"model.visual.blocks.13.mlp.linear_fc1.weight": "model.safetensors",
|
| 498 |
+
"model.visual.blocks.13.mlp.linear_fc2.bias": "model.safetensors",
|
| 499 |
+
"model.visual.blocks.13.mlp.linear_fc2.weight": "model.safetensors",
|
| 500 |
+
"model.visual.blocks.13.norm1.bias": "model.safetensors",
|
| 501 |
+
"model.visual.blocks.13.norm1.weight": "model.safetensors",
|
| 502 |
+
"model.visual.blocks.13.norm2.bias": "model.safetensors",
|
| 503 |
+
"model.visual.blocks.13.norm2.weight": "model.safetensors",
|
| 504 |
+
"model.visual.blocks.14.attn.proj.bias": "model.safetensors",
|
| 505 |
+
"model.visual.blocks.14.attn.proj.weight": "model.safetensors",
|
| 506 |
+
"model.visual.blocks.14.attn.qkv.bias": "model.safetensors",
|
| 507 |
+
"model.visual.blocks.14.attn.qkv.weight": "model.safetensors",
|
| 508 |
+
"model.visual.blocks.14.mlp.linear_fc1.bias": "model.safetensors",
|
| 509 |
+
"model.visual.blocks.14.mlp.linear_fc1.weight": "model.safetensors",
|
| 510 |
+
"model.visual.blocks.14.mlp.linear_fc2.bias": "model.safetensors",
|
| 511 |
+
"model.visual.blocks.14.mlp.linear_fc2.weight": "model.safetensors",
|
| 512 |
+
"model.visual.blocks.14.norm1.bias": "model.safetensors",
|
| 513 |
+
"model.visual.blocks.14.norm1.weight": "model.safetensors",
|
| 514 |
+
"model.visual.blocks.14.norm2.bias": "model.safetensors",
|
| 515 |
+
"model.visual.blocks.14.norm2.weight": "model.safetensors",
|
| 516 |
+
"model.visual.blocks.15.attn.proj.bias": "model.safetensors",
|
| 517 |
+
"model.visual.blocks.15.attn.proj.weight": "model.safetensors",
|
| 518 |
+
"model.visual.blocks.15.attn.qkv.bias": "model.safetensors",
|
| 519 |
+
"model.visual.blocks.15.attn.qkv.weight": "model.safetensors",
|
| 520 |
+
"model.visual.blocks.15.mlp.linear_fc1.bias": "model.safetensors",
|
| 521 |
+
"model.visual.blocks.15.mlp.linear_fc1.weight": "model.safetensors",
|
| 522 |
+
"model.visual.blocks.15.mlp.linear_fc2.bias": "model.safetensors",
|
| 523 |
+
"model.visual.blocks.15.mlp.linear_fc2.weight": "model.safetensors",
|
| 524 |
+
"model.visual.blocks.15.norm1.bias": "model.safetensors",
|
| 525 |
+
"model.visual.blocks.15.norm1.weight": "model.safetensors",
|
| 526 |
+
"model.visual.blocks.15.norm2.bias": "model.safetensors",
|
| 527 |
+
"model.visual.blocks.15.norm2.weight": "model.safetensors",
|
| 528 |
+
"model.visual.blocks.16.attn.proj.bias": "model.safetensors",
|
| 529 |
+
"model.visual.blocks.16.attn.proj.weight": "model.safetensors",
|
| 530 |
+
"model.visual.blocks.16.attn.qkv.bias": "model.safetensors",
|
| 531 |
+
"model.visual.blocks.16.attn.qkv.weight": "model.safetensors",
|
| 532 |
+
"model.visual.blocks.16.mlp.linear_fc1.bias": "model.safetensors",
|
| 533 |
+
"model.visual.blocks.16.mlp.linear_fc1.weight": "model.safetensors",
|
| 534 |
+
"model.visual.blocks.16.mlp.linear_fc2.bias": "model.safetensors",
|
| 535 |
+
"model.visual.blocks.16.mlp.linear_fc2.weight": "model.safetensors",
|
| 536 |
+
"model.visual.blocks.16.norm1.bias": "model.safetensors",
|
| 537 |
+
"model.visual.blocks.16.norm1.weight": "model.safetensors",
|
| 538 |
+
"model.visual.blocks.16.norm2.bias": "model.safetensors",
|
| 539 |
+
"model.visual.blocks.16.norm2.weight": "model.safetensors",
|
| 540 |
+
"model.visual.blocks.17.attn.proj.bias": "model.safetensors",
|
| 541 |
+
"model.visual.blocks.17.attn.proj.weight": "model.safetensors",
|
| 542 |
+
"model.visual.blocks.17.attn.qkv.bias": "model.safetensors",
|
| 543 |
+
"model.visual.blocks.17.attn.qkv.weight": "model.safetensors",
|
| 544 |
+
"model.visual.blocks.17.mlp.linear_fc1.bias": "model.safetensors",
|
| 545 |
+
"model.visual.blocks.17.mlp.linear_fc1.weight": "model.safetensors",
|
| 546 |
+
"model.visual.blocks.17.mlp.linear_fc2.bias": "model.safetensors",
|
| 547 |
+
"model.visual.blocks.17.mlp.linear_fc2.weight": "model.safetensors",
|
| 548 |
+
"model.visual.blocks.17.norm1.bias": "model.safetensors",
|
| 549 |
+
"model.visual.blocks.17.norm1.weight": "model.safetensors",
|
| 550 |
+
"model.visual.blocks.17.norm2.bias": "model.safetensors",
|
| 551 |
+
"model.visual.blocks.17.norm2.weight": "model.safetensors",
|
| 552 |
+
"model.visual.blocks.18.attn.proj.bias": "model.safetensors",
|
| 553 |
+
"model.visual.blocks.18.attn.proj.weight": "model.safetensors",
|
| 554 |
+
"model.visual.blocks.18.attn.qkv.bias": "model.safetensors",
|
| 555 |
+
"model.visual.blocks.18.attn.qkv.weight": "model.safetensors",
|
| 556 |
+
"model.visual.blocks.18.mlp.linear_fc1.bias": "model.safetensors",
|
| 557 |
+
"model.visual.blocks.18.mlp.linear_fc1.weight": "model.safetensors",
|
| 558 |
+
"model.visual.blocks.18.mlp.linear_fc2.bias": "model.safetensors",
|
| 559 |
+
"model.visual.blocks.18.mlp.linear_fc2.weight": "model.safetensors",
|
| 560 |
+
"model.visual.blocks.18.norm1.bias": "model.safetensors",
|
| 561 |
+
"model.visual.blocks.18.norm1.weight": "model.safetensors",
|
| 562 |
+
"model.visual.blocks.18.norm2.bias": "model.safetensors",
|
| 563 |
+
"model.visual.blocks.18.norm2.weight": "model.safetensors",
|
| 564 |
+
"model.visual.blocks.19.attn.proj.bias": "model.safetensors",
|
| 565 |
+
"model.visual.blocks.19.attn.proj.weight": "model.safetensors",
|
| 566 |
+
"model.visual.blocks.19.attn.qkv.bias": "model.safetensors",
|
| 567 |
+
"model.visual.blocks.19.attn.qkv.weight": "model.safetensors",
|
| 568 |
+
"model.visual.blocks.19.mlp.linear_fc1.bias": "model.safetensors",
|
| 569 |
+
"model.visual.blocks.19.mlp.linear_fc1.weight": "model.safetensors",
|
| 570 |
+
"model.visual.blocks.19.mlp.linear_fc2.bias": "model.safetensors",
|
| 571 |
+
"model.visual.blocks.19.mlp.linear_fc2.weight": "model.safetensors",
|
| 572 |
+
"model.visual.blocks.19.norm1.bias": "model.safetensors",
|
| 573 |
+
"model.visual.blocks.19.norm1.weight": "model.safetensors",
|
| 574 |
+
"model.visual.blocks.19.norm2.bias": "model.safetensors",
|
| 575 |
+
"model.visual.blocks.19.norm2.weight": "model.safetensors",
|
| 576 |
+
"model.visual.blocks.2.attn.proj.bias": "model.safetensors",
|
| 577 |
+
"model.visual.blocks.2.attn.proj.weight": "model.safetensors",
|
| 578 |
+
"model.visual.blocks.2.attn.qkv.bias": "model.safetensors",
|
| 579 |
+
"model.visual.blocks.2.attn.qkv.weight": "model.safetensors",
|
| 580 |
+
"model.visual.blocks.2.mlp.linear_fc1.bias": "model.safetensors",
|
| 581 |
+
"model.visual.blocks.2.mlp.linear_fc1.weight": "model.safetensors",
|
| 582 |
+
"model.visual.blocks.2.mlp.linear_fc2.bias": "model.safetensors",
|
| 583 |
+
"model.visual.blocks.2.mlp.linear_fc2.weight": "model.safetensors",
|
| 584 |
+
"model.visual.blocks.2.norm1.bias": "model.safetensors",
|
| 585 |
+
"model.visual.blocks.2.norm1.weight": "model.safetensors",
|
| 586 |
+
"model.visual.blocks.2.norm2.bias": "model.safetensors",
|
| 587 |
+
"model.visual.blocks.2.norm2.weight": "model.safetensors",
|
| 588 |
+
"model.visual.blocks.20.attn.proj.bias": "model.safetensors",
|
| 589 |
+
"model.visual.blocks.20.attn.proj.weight": "model.safetensors",
|
| 590 |
+
"model.visual.blocks.20.attn.qkv.bias": "model.safetensors",
|
| 591 |
+
"model.visual.blocks.20.attn.qkv.weight": "model.safetensors",
|
| 592 |
+
"model.visual.blocks.20.mlp.linear_fc1.bias": "model.safetensors",
|
| 593 |
+
"model.visual.blocks.20.mlp.linear_fc1.weight": "model.safetensors",
|
| 594 |
+
"model.visual.blocks.20.mlp.linear_fc2.bias": "model.safetensors",
|
| 595 |
+
"model.visual.blocks.20.mlp.linear_fc2.weight": "model.safetensors",
|
| 596 |
+
"model.visual.blocks.20.norm1.bias": "model.safetensors",
|
| 597 |
+
"model.visual.blocks.20.norm1.weight": "model.safetensors",
|
| 598 |
+
"model.visual.blocks.20.norm2.bias": "model.safetensors",
|
| 599 |
+
"model.visual.blocks.20.norm2.weight": "model.safetensors",
|
| 600 |
+
"model.visual.blocks.21.attn.proj.bias": "model.safetensors",
|
| 601 |
+
"model.visual.blocks.21.attn.proj.weight": "model.safetensors",
|
| 602 |
+
"model.visual.blocks.21.attn.qkv.bias": "model.safetensors",
|
| 603 |
+
"model.visual.blocks.21.attn.qkv.weight": "model.safetensors",
|
| 604 |
+
"model.visual.blocks.21.mlp.linear_fc1.bias": "model.safetensors",
|
| 605 |
+
"model.visual.blocks.21.mlp.linear_fc1.weight": "model.safetensors",
|
| 606 |
+
"model.visual.blocks.21.mlp.linear_fc2.bias": "model.safetensors",
|
| 607 |
+
"model.visual.blocks.21.mlp.linear_fc2.weight": "model.safetensors",
|
| 608 |
+
"model.visual.blocks.21.norm1.bias": "model.safetensors",
|
| 609 |
+
"model.visual.blocks.21.norm1.weight": "model.safetensors",
|
| 610 |
+
"model.visual.blocks.21.norm2.bias": "model.safetensors",
|
| 611 |
+
"model.visual.blocks.21.norm2.weight": "model.safetensors",
|
| 612 |
+
"model.visual.blocks.22.attn.proj.bias": "model.safetensors",
|
| 613 |
+
"model.visual.blocks.22.attn.proj.weight": "model.safetensors",
|
| 614 |
+
"model.visual.blocks.22.attn.qkv.bias": "model.safetensors",
|
| 615 |
+
"model.visual.blocks.22.attn.qkv.weight": "model.safetensors",
|
| 616 |
+
"model.visual.blocks.22.mlp.linear_fc1.bias": "model.safetensors",
|
| 617 |
+
"model.visual.blocks.22.mlp.linear_fc1.weight": "model.safetensors",
|
| 618 |
+
"model.visual.blocks.22.mlp.linear_fc2.bias": "model.safetensors",
|
| 619 |
+
"model.visual.blocks.22.mlp.linear_fc2.weight": "model.safetensors",
|
| 620 |
+
"model.visual.blocks.22.norm1.bias": "model.safetensors",
|
| 621 |
+
"model.visual.blocks.22.norm1.weight": "model.safetensors",
|
| 622 |
+
"model.visual.blocks.22.norm2.bias": "model.safetensors",
|
| 623 |
+
"model.visual.blocks.22.norm2.weight": "model.safetensors",
|
| 624 |
+
"model.visual.blocks.23.attn.proj.bias": "model.safetensors",
|
| 625 |
+
"model.visual.blocks.23.attn.proj.weight": "model.safetensors",
|
| 626 |
+
"model.visual.blocks.23.attn.qkv.bias": "model.safetensors",
|
| 627 |
+
"model.visual.blocks.23.attn.qkv.weight": "model.safetensors",
|
| 628 |
+
"model.visual.blocks.23.mlp.linear_fc1.bias": "model.safetensors",
|
| 629 |
+
"model.visual.blocks.23.mlp.linear_fc1.weight": "model.safetensors",
|
| 630 |
+
"model.visual.blocks.23.mlp.linear_fc2.bias": "model.safetensors",
|
| 631 |
+
"model.visual.blocks.23.mlp.linear_fc2.weight": "model.safetensors",
|
| 632 |
+
"model.visual.blocks.23.norm1.bias": "model.safetensors",
|
| 633 |
+
"model.visual.blocks.23.norm1.weight": "model.safetensors",
|
| 634 |
+
"model.visual.blocks.23.norm2.bias": "model.safetensors",
|
| 635 |
+
"model.visual.blocks.23.norm2.weight": "model.safetensors",
|
| 636 |
+
"model.visual.blocks.3.attn.proj.bias": "model.safetensors",
|
| 637 |
+
"model.visual.blocks.3.attn.proj.weight": "model.safetensors",
|
| 638 |
+
"model.visual.blocks.3.attn.qkv.bias": "model.safetensors",
|
| 639 |
+
"model.visual.blocks.3.attn.qkv.weight": "model.safetensors",
|
| 640 |
+
"model.visual.blocks.3.mlp.linear_fc1.bias": "model.safetensors",
|
| 641 |
+
"model.visual.blocks.3.mlp.linear_fc1.weight": "model.safetensors",
|
| 642 |
+
"model.visual.blocks.3.mlp.linear_fc2.bias": "model.safetensors",
|
| 643 |
+
"model.visual.blocks.3.mlp.linear_fc2.weight": "model.safetensors",
|
| 644 |
+
"model.visual.blocks.3.norm1.bias": "model.safetensors",
|
| 645 |
+
"model.visual.blocks.3.norm1.weight": "model.safetensors",
|
| 646 |
+
"model.visual.blocks.3.norm2.bias": "model.safetensors",
|
| 647 |
+
"model.visual.blocks.3.norm2.weight": "model.safetensors",
|
| 648 |
+
"model.visual.blocks.4.attn.proj.bias": "model.safetensors",
|
| 649 |
+
"model.visual.blocks.4.attn.proj.weight": "model.safetensors",
|
| 650 |
+
"model.visual.blocks.4.attn.qkv.bias": "model.safetensors",
|
| 651 |
+
"model.visual.blocks.4.attn.qkv.weight": "model.safetensors",
|
| 652 |
+
"model.visual.blocks.4.mlp.linear_fc1.bias": "model.safetensors",
|
| 653 |
+
"model.visual.blocks.4.mlp.linear_fc1.weight": "model.safetensors",
|
| 654 |
+
"model.visual.blocks.4.mlp.linear_fc2.bias": "model.safetensors",
|
| 655 |
+
"model.visual.blocks.4.mlp.linear_fc2.weight": "model.safetensors",
|
| 656 |
+
"model.visual.blocks.4.norm1.bias": "model.safetensors",
|
| 657 |
+
"model.visual.blocks.4.norm1.weight": "model.safetensors",
|
| 658 |
+
"model.visual.blocks.4.norm2.bias": "model.safetensors",
|
| 659 |
+
"model.visual.blocks.4.norm2.weight": "model.safetensors",
|
| 660 |
+
"model.visual.blocks.5.attn.proj.bias": "model.safetensors",
|
| 661 |
+
"model.visual.blocks.5.attn.proj.weight": "model.safetensors",
|
| 662 |
+
"model.visual.blocks.5.attn.qkv.bias": "model.safetensors",
|
| 663 |
+
"model.visual.blocks.5.attn.qkv.weight": "model.safetensors",
|
| 664 |
+
"model.visual.blocks.5.mlp.linear_fc1.bias": "model.safetensors",
|
| 665 |
+
"model.visual.blocks.5.mlp.linear_fc1.weight": "model.safetensors",
|
| 666 |
+
"model.visual.blocks.5.mlp.linear_fc2.bias": "model.safetensors",
|
| 667 |
+
"model.visual.blocks.5.mlp.linear_fc2.weight": "model.safetensors",
|
| 668 |
+
"model.visual.blocks.5.norm1.bias": "model.safetensors",
|
| 669 |
+
"model.visual.blocks.5.norm1.weight": "model.safetensors",
|
| 670 |
+
"model.visual.blocks.5.norm2.bias": "model.safetensors",
|
| 671 |
+
"model.visual.blocks.5.norm2.weight": "model.safetensors",
|
| 672 |
+
"model.visual.blocks.6.attn.proj.bias": "model.safetensors",
|
| 673 |
+
"model.visual.blocks.6.attn.proj.weight": "model.safetensors",
|
| 674 |
+
"model.visual.blocks.6.attn.qkv.bias": "model.safetensors",
|
| 675 |
+
"model.visual.blocks.6.attn.qkv.weight": "model.safetensors",
|
| 676 |
+
"model.visual.blocks.6.mlp.linear_fc1.bias": "model.safetensors",
|
| 677 |
+
"model.visual.blocks.6.mlp.linear_fc1.weight": "model.safetensors",
|
| 678 |
+
"model.visual.blocks.6.mlp.linear_fc2.bias": "model.safetensors",
|
| 679 |
+
"model.visual.blocks.6.mlp.linear_fc2.weight": "model.safetensors",
|
| 680 |
+
"model.visual.blocks.6.norm1.bias": "model.safetensors",
|
| 681 |
+
"model.visual.blocks.6.norm1.weight": "model.safetensors",
|
| 682 |
+
"model.visual.blocks.6.norm2.bias": "model.safetensors",
|
| 683 |
+
"model.visual.blocks.6.norm2.weight": "model.safetensors",
|
| 684 |
+
"model.visual.blocks.7.attn.proj.bias": "model.safetensors",
|
| 685 |
+
"model.visual.blocks.7.attn.proj.weight": "model.safetensors",
|
| 686 |
+
"model.visual.blocks.7.attn.qkv.bias": "model.safetensors",
|
| 687 |
+
"model.visual.blocks.7.attn.qkv.weight": "model.safetensors",
|
| 688 |
+
"model.visual.blocks.7.mlp.linear_fc1.bias": "model.safetensors",
|
| 689 |
+
"model.visual.blocks.7.mlp.linear_fc1.weight": "model.safetensors",
|
| 690 |
+
"model.visual.blocks.7.mlp.linear_fc2.bias": "model.safetensors",
|
| 691 |
+
"model.visual.blocks.7.mlp.linear_fc2.weight": "model.safetensors",
|
| 692 |
+
"model.visual.blocks.7.norm1.bias": "model.safetensors",
|
| 693 |
+
"model.visual.blocks.7.norm1.weight": "model.safetensors",
|
| 694 |
+
"model.visual.blocks.7.norm2.bias": "model.safetensors",
|
| 695 |
+
"model.visual.blocks.7.norm2.weight": "model.safetensors",
|
| 696 |
+
"model.visual.blocks.8.attn.proj.bias": "model.safetensors",
|
| 697 |
+
"model.visual.blocks.8.attn.proj.weight": "model.safetensors",
|
| 698 |
+
"model.visual.blocks.8.attn.qkv.bias": "model.safetensors",
|
| 699 |
+
"model.visual.blocks.8.attn.qkv.weight": "model.safetensors",
|
| 700 |
+
"model.visual.blocks.8.mlp.linear_fc1.bias": "model.safetensors",
|
| 701 |
+
"model.visual.blocks.8.mlp.linear_fc1.weight": "model.safetensors",
|
| 702 |
+
"model.visual.blocks.8.mlp.linear_fc2.bias": "model.safetensors",
|
| 703 |
+
"model.visual.blocks.8.mlp.linear_fc2.weight": "model.safetensors",
|
| 704 |
+
"model.visual.blocks.8.norm1.bias": "model.safetensors",
|
| 705 |
+
"model.visual.blocks.8.norm1.weight": "model.safetensors",
|
| 706 |
+
"model.visual.blocks.8.norm2.bias": "model.safetensors",
|
| 707 |
+
"model.visual.blocks.8.norm2.weight": "model.safetensors",
|
| 708 |
+
"model.visual.blocks.9.attn.proj.bias": "model.safetensors",
|
| 709 |
+
"model.visual.blocks.9.attn.proj.weight": "model.safetensors",
|
| 710 |
+
"model.visual.blocks.9.attn.qkv.bias": "model.safetensors",
|
| 711 |
+
"model.visual.blocks.9.attn.qkv.weight": "model.safetensors",
|
| 712 |
+
"model.visual.blocks.9.mlp.linear_fc1.bias": "model.safetensors",
|
| 713 |
+
"model.visual.blocks.9.mlp.linear_fc1.weight": "model.safetensors",
|
| 714 |
+
"model.visual.blocks.9.mlp.linear_fc2.bias": "model.safetensors",
|
| 715 |
+
"model.visual.blocks.9.mlp.linear_fc2.weight": "model.safetensors",
|
| 716 |
+
"model.visual.blocks.9.norm1.bias": "model.safetensors",
|
| 717 |
+
"model.visual.blocks.9.norm1.weight": "model.safetensors",
|
| 718 |
+
"model.visual.blocks.9.norm2.bias": "model.safetensors",
|
| 719 |
+
"model.visual.blocks.9.norm2.weight": "model.safetensors",
|
| 720 |
+
"model.visual.merger.linear_fc1.bias": "model.safetensors",
|
| 721 |
+
"model.visual.merger.linear_fc1.weight": "model.safetensors",
|
| 722 |
+
"model.visual.merger.linear_fc2.bias": "model.safetensors",
|
| 723 |
+
"model.visual.merger.linear_fc2.weight": "model.safetensors",
|
| 724 |
+
"model.visual.merger.norm.bias": "model.safetensors",
|
| 725 |
+
"model.visual.merger.norm.weight": "model.safetensors",
|
| 726 |
+
"model.visual.patch_embed.proj.bias": "model.safetensors",
|
| 727 |
+
"model.visual.patch_embed.proj.weight": "model.safetensors",
|
| 728 |
+
"model.visual.pos_embed.weight": "model.safetensors",
|
| 729 |
+
"mtp.fc.weight": "model_mtp.safetensors",
|
| 730 |
+
"mtp.layers.0.input_layernorm.weight": "model_mtp.safetensors",
|
| 731 |
+
"mtp.layers.0.mlp.down_proj.weight": "model_mtp.safetensors",
|
| 732 |
+
"mtp.layers.0.mlp.gate_proj.weight": "model_mtp.safetensors",
|
| 733 |
+
"mtp.layers.0.mlp.up_proj.weight": "model_mtp.safetensors",
|
| 734 |
+
"mtp.layers.0.post_attention_layernorm.weight": "model_mtp.safetensors",
|
| 735 |
+
"mtp.layers.0.self_attn.k_norm.weight": "model_mtp.safetensors",
|
| 736 |
+
"mtp.layers.0.self_attn.k_proj.weight": "model_mtp.safetensors",
|
| 737 |
+
"mtp.layers.0.self_attn.o_proj.weight": "model_mtp.safetensors",
|
| 738 |
+
"mtp.layers.0.self_attn.q_norm.weight": "model_mtp.safetensors",
|
| 739 |
+
"mtp.layers.0.self_attn.q_proj.weight": "model_mtp.safetensors",
|
| 740 |
+
"mtp.layers.0.self_attn.v_proj.weight": "model_mtp.safetensors",
|
| 741 |
+
"mtp.norm.weight": "model_mtp.safetensors",
|
| 742 |
+
"mtp.pre_fc_norm_embedding.weight": "model_mtp.safetensors",
|
| 743 |
+
"mtp.pre_fc_norm_hidden.weight": "model_mtp.safetensors"
|
| 744 |
+
}
|
| 745 |
+
}
|
model_mtp.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7f993d7b896c6d3c72ee66fd446b28bcf316d5f5ce4a0427c0442dfe461cbe1b
|
| 3 |
+
size 241200704
|
ocr_benchmark_specialized.png
ADDED
|
Git LFS Details
|
ocr_preferences.svg
ADDED
|
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 |
+
}
|
st.svg
ADDED
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:87a7830d63fcf43bf241c3c5242e96e62dd3fdc29224ca26fed8ea333db72de4
|
| 3 |
+
size 19989343
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": false,
|
| 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": "TokenizersBackend",
|
| 29 |
+
"unk_token": null,
|
| 30 |
+
"video_token": "<|video_pad|>",
|
| 31 |
+
"vision_bos_token": "<|vision_start|>",
|
| 32 |
+
"vision_eos_token": "<|vision_end|>"
|
| 33 |
+
}
|