Duplicate from athrael-soju/UltraVBERT

Co-authored-by: Athrael Soju <athrael-soju@users.noreply.huggingface.co>

.gitattributes

@@ -0,0 +1,36 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+ultravbert-hero.png filter=lfs diff=lfs merge=lfs -text

LICENSE

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

LICENSE-APACHE

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

README.md

@@ -0,0 +1,164 @@
+---
+license: mit
+language:
+- en
+base_model:
+- dmis-lab/UltraBERT-150M
+base_model_relation: finetune
+datasets:
+- HuggingFaceM4/Docmatix
+- HuggingFaceM4/FineVision
+- HuggingFaceM4/the_cauldron
+- rlhn/rlhn-680K
+tags:
+- ultravbert
+- ultrabert
+- vision-language
+- multimodal
+- document-retrieval
+- colpali
+- vidore
+pipeline_tag: visual-document-retrieval
+library_name: transformers
+---
+
+<img src="ultravbert-hero.png" width="100%" alt="UltraVBERT">
+
+# UltraVBERT
+
+UltraVBERT is a 293M-parameter vision-language encoder built as a base for visual document retrievers. It combines the
+[UltraBERT-150M](https://huggingface.co/dmis-lab/UltraBERT-150M) text encoder with the
+[SigLIP2-B/16-512](https://huggingface.co/google/siglip2-base-patch16-512) vision encoder and produces a contextual
+embedding for every text and image token.
+
+## Models
+
+<table style="display: table; width: 100%;">
+<thead>
+<tr><th align="left">Model</th><th align="left">Role</th><th align="left">Status</th></tr>
+</thead>
+<tbody>
+<tr><td><a href="https://huggingface.co/athrael-soju/UltraVBERT">UltraVBERT</a></td><td>Base encoder aligned to page images</td><td>Available</td></tr>
+<tr><td><a href="https://huggingface.co/athrael-soju/BiUltraVBERT">BiUltraVBERT</a></td><td>Single-vector retriever for visual documents</td><td>Available</td></tr>
+<tr><td><a href="https://huggingface.co/athrael-soju/ColUltraVBERT">ColUltraVBERT</a></td><td>Late-interaction retriever for visual documents</td><td>Available · <a href="https://huggingface.co/spaces/athrael-soju/ColUltraVBERT">Demo</a></td></tr>
+</tbody>
+</table>
+
+## Architecture
+
+SigLIP2 encodes each 512 px tile of a page into 1,024 patches. The connector merges each 4 x 4 block of patches into
+one token, giving 64 tokens per tile, then projects each token to UltraBERT's hidden size and RMS-normalizes it.
+UltraBERT then attends over image and text tokens together.
+
+UltraVBERT builds on [ModernVBERT](https://huggingface.co/ModernVBERT/modernvbert)
+([arXiv 2510.01149](https://arxiv.org/abs/2510.01149)). It replaces the Ettin-150M text encoder with UltraBERT
+([arXiv 2610.15028](https://arxiv.org/abs/2610.15028)), normalizes the connector output, raises the page limit from
+2,048 px and 16 tiles to 3,072 px and 27 tiles, and adds a training stage on rendered text pages.
+
+| Component | Configuration | Parameters |
+| --- | --- | ---: |
+| Text encoder | UltraBERT-150M: 24 layers, hidden size 768, 6 heads; local attention over ±128 tokens in 2 of every 3 layers; 40 added special tokens | 151,130,880 |
+| Vision encoder | SigLIP2-B/16-512: 12 layers, 16 px patches, 512 px tiles | 93,520,128 |
+| Connector | 4 x 4 pixel shuffle, linear 12,288 to 768, RMSNorm | 9,437,952 |
+| MLM head | UltraBERT's released head: two 768 x 768 layers and an output layer extended to 49,280 tokens | 39,026,688 |
+| Total | | 293,115,648 |
+
+The encoder without the MLM head has 254M parameters. Weights are stored in bfloat16.
+
+The processor scales a page to 3,072 px on its longest edge, reduces that in 512 px steps until the page fits in 27
+tiles, and adds a downscaled view of the whole page. A grid holds at most six tiles per side, so the largest that
+fits is 5 x 5, for at most 26 x 64 = 1,664 image tokens per page. For shorter
+sequences, set
+`processor.image_processor.size = {"longest_edge": 2048}` (at most 1,088 image tokens) or lower
+`processor.image_processor.max_image_tiles`.
+
+## Usage
+
+Requires `transformers>=5.20`.
+
+```python
+import torch
+from PIL import Image
+from transformers import AutoModel, AutoModelForMaskedLM, AutoProcessor
+
+processor = AutoProcessor.from_pretrained("athrael-soju/UltraVBERT", trust_remote_code=True)
+page = Image.open("page.png")
+
+# Predict a masked word from the page.
+model = AutoModelForMaskedLM.from_pretrained(
+    "athrael-soju/UltraVBERT", trust_remote_code=True, dtype=torch.bfloat16, device_map="auto"
+)
+inputs = processor(text="<image>The table reports revenue in <|mask|>.", images=[page], return_tensors="pt").to(model.device)
+with torch.no_grad():
+    logits = model(**inputs).logits
+masked_logits = logits[inputs.input_ids == processor.tokenizer.mask_token_id]
+print(processor.tokenizer.batch_decode(masked_logits.argmax(dim=-1)[:, None]))
+
+# One 768-d embedding per text and image token.
+encoder = AutoModel.from_pretrained(
+    "athrael-soju/UltraVBERT", trust_remote_code=True, dtype=torch.bfloat16, device_map="auto"
+)
+inputs = processor(text="<image>", images=[page], return_tensors="pt").to(encoder.device)
+with torch.no_grad():
+    states = encoder(**inputs).last_hidden_state
+```
+
+`AutoModelForSequenceClassification` and `AutoModelForTokenClassification` add task heads on the same encoder.
+
+## Training
+
+UltraVBERT is trained with masked language modelling on the answers in image-grounded conversations, with half of the
+answer tokens masked (80% `<|mask|>`, 10% random token, 10% unchanged).
+
+- DoRA adapters (rank 64, alpha 16) on both encoders and the MLM head, peak learning rate 1e-4
+- Connector trained in full at 1e-3; image tokens start at 10x the RMS of UltraBERT's word embeddings
+- The 40 added token embeddings trained directly
+- AdamW (beta2 0.95), weight decay 0.1, gradient clipping 1.0, bfloat16 autocast
+
+| Stage | Steps | Text tokens | Data |
+| --- | ---: | ---: | --- |
+| Reading | 3,600 | 0.36B | Docmatix pages and rendered RLHN-680K passages, half each |
+| Mixture | 34,206 | 8.9B | The mix below |
+
+The mixture stage uses a warmup-stable-decay schedule with a 1 - sqrt decay over the last 20% of steps. Pages are
+resized to a 1,024 px longest edge until the decay begins, then to random longest edges from 1,024 to 3,072 px.
+
+| Category | Share of text tokens | Sources |
+| --- | ---: | --- |
+| Text instructions | 48.18% | 16 FineVision text subsets |
+| OCR and documents | 16.21% | Docmatix and 9 FineVision subsets |
+| Screenshot to code | 16.71% | 4 FineVision subsets |
+| Rendered text pages | 5.00% | RLHN-680K passages |
+| Charts | 5.00% | 8 FineVision subsets |
+| Captioning | 3.13% | 2 FineVision subsets |
+| Tables | 3.00% | 6 FineVision subsets |
+| Reasoning and math | 1.61% | 7 FineVision subsets |
+| Real-world VQA | 1.16% | 2 FineVision subsets and The Cauldron's TallyQA |
+
+Datasets: [Docmatix](https://huggingface.co/datasets/HuggingFaceM4/Docmatix),
+[FineVision](https://huggingface.co/datasets/HuggingFaceM4/FineVision),
+[The Cauldron](https://huggingface.co/datasets/HuggingFaceM4/the_cauldron) and
+[RLHN-680K](https://huggingface.co/datasets/rlhn/rlhn-680K). Subsets used:
+
+| Category | Subsets |
+| --- | --- |
+| Text instructions | `text_code_feedback`, `text_codefeedback_filtered_instruction`, `text_infinitymath`, `text_mathinstruct`, `text_mathqa`, `text_mathstepdpo10k`, `text_numinamath_cot`, `text_openhermes_2_5`, `text_OpenMathInstruct-2`, `text_openorca`, `text_orcamath`, `text_pythoncode25k`, `text_pythoncodealpaca`, `text_ruozhiba`, `text_theoremqa`, `text_wizardlm_evol` |
+| OCR and documents | Docmatix, `olmOCR-mix-0225-documents`, `CoSyn_400k_document`, `CoSyn_400k_diagram`, `SynthFormulaNet`, `latexformulas`, `k12_printing`, `tal_ocr_eng`, `ocrvqa`, `DoclingMatix` |
+| Screenshot to code | `SynthCodeNet`, `datik`, `datikz`, `websight` |
+| Charts | `CoSyn_400k_chart`, `Unichart`, `SynthChartNet`, `figureqa`, `mapqa`, `chart2text`, `mmc_instruct`, `plotqa` |
+| Captioning | `densefusion_1m`, `localized_narratives` |
+| Tables | `CoSyn_400k_table`, `vqaonbd`, `robut_wikisql`, `robut_wtq`, `finqa`, `robut_sqa` |
+| Reasoning and math | `CoSyn_400k_math`, `mavis_math_metagen`, `geomverse`, `geo170k(qa)`, `geo170k(align)`, `CoSyn_400k_circuit`, `CoSyn_400k_chemical` |
+| Real-world VQA | `CoSyn_400k_graphic`, `mmevol`, The Cauldron `tallyqa` |
+
+## Limits
+
+- English only.
+- Trained on inputs of up to 7,000 tokens and 3 images.
+- Pages larger than the tile cap are downscaled, which can blur small print.
+
+## License
+
+`MIT`. The UltraBERT-150M and SigLIP2-B/16-512 components are Apache-2.0, with their license text in
+`LICENSE-APACHE`. Training used passages from
+RLHN-680K (CC-BY-SA-4.0).

config.json

@@ -0,0 +1,103 @@
+{
+  "architectures": [
+    "UltraVBertForMaskedLM"
+  ],
+  "auto_map": {
+    "AutoConfig": "configuration_ultravbert.UltraVBertConfig",
+    "AutoModel": "modeling_ultravbert.UltraVBertModel",
+    "AutoModelForMaskedLM": "modeling_ultravbert.UltraVBertForMaskedLM",
+    "AutoModelForSequenceClassification": "modeling_ultravbert.UltraVBertForSequenceClassification",
+    "AutoModelForTokenClassification": "modeling_ultravbert.UltraVBertForTokenClassification"
+  },
+  "classifier_dropout": 0.0,
+  "classifier_pooling": "cls",
+  "connector_norm": true,
+  "dtype": "bfloat16",
+  "image_token_id": 49191,
+  "initializer_cutoff_factor": 2.0,
+  "initializer_range": 0.02,
+  "model_type": "ultravbert",
+  "pixel_shuffle_factor": 4,
+  "text_config": {
+    "architectures": [
+      "UltrabertForMaskedLM"
+    ],
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "bos_token_id": 0,
+    "classifier_dropout": 0.0,
+    "classifier_pooling": "cls",
+    "dtype": "float32",
+    "eos_token_id": 0,
+    "head_dim": 128,
+    "hidden_act": "silu",
+    "hidden_size": 768,
+    "initializer_range": 0.02,
+    "intermediate_size": 1024,
+    "layer_types": [
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention"
+    ],
+    "max_position_embeddings": 32768,
+    "mlp_bias": false,
+    "model_type": "ultrabert",
+    "num_attention_heads": 6,
+    "num_hidden_layers": 24,
+    "num_key_value_heads": 6,
+    "pad_token_id": 0,
+    "prediction_head_size": 768,
+    "rms_norm_eps": 1e-06,
+    "rope_parameters": {
+      "full_attention": {
+        "rope_theta": 2560000.0,
+        "rope_type": "default"
+      },
+      "sliding_attention": {
+        "rope_theta": 10000.0,
+        "rope_type": "default"
+      }
+    },
+    "sliding_window": 128,
+    "tie_word_embeddings": false,
+    "use_cache": false,
+    "vocab_size": 49280
+  },
+  "tie_word_embeddings": false,
+  "transformers_version": "5.21.0.dev0",
+  "vision_config": {
+    "attention_dropout": 0.0,
+    "hidden_act": "gelu_pytorch_tanh",
+    "hidden_size": 768,
+    "image_size": 512,
+    "intermediate_size": 3072,
+    "layer_norm_eps": 1e-06,
+    "model_type": "siglip_vision_model",
+    "num_attention_heads": 12,
+    "num_channels": 3,
+    "num_hidden_layers": 12,
+    "patch_size": 16,
+    "vision_use_head": false
+  }
+}

configuration_ultravbert.py

@@ -0,0 +1,29 @@
+from typing import Literal
+
+from huggingface_hub.dataclasses import strict
+from transformers import AutoConfig
+from transformers.configuration_utils import PreTrainedConfig, SubConfigSpec
+
+
+@strict
+class UltraVBertConfig(PreTrainedConfig):
+    model_type = "ultravbert"
+    sub_configs_defaults = {
+        "text_config": SubConfigSpec(config_class=AutoConfig, model_type="ultrabert"),
+        "vision_config": SubConfigSpec(config_class=AutoConfig, model_type="siglip_vision_model"),
+    }
+
+    text_config: PreTrainedConfig | dict | None = None
+    vision_config: PreTrainedConfig | dict | None = None
+
+    image_token_id: int = 49191
+    pixel_shuffle_factor: int = 4
+    initializer_range: float = 0.02
+    initializer_cutoff_factor: float = 2.0
+    classifier_pooling: Literal["cls", "mean"] = "cls"
+    classifier_dropout: float | int = 0.0
+    tie_word_embeddings: bool = False
+    connector_norm: bool = True
+
+
+__all__ = ["UltraVBertConfig"]

image_processing_ultravbert.py

@@ -0,0 +1,73 @@
+import math
+
+from transformers.image_utils import SizeDict
+from transformers.models.idefics3.image_processing_idefics3 import (
+    MAX_IMAGE_SIZE,
+    Idefics3ImageProcessor,
+    Idefics3ImageProcessorKwargs,
+    _resize_output_size_rescale_to_max_len,
+    _resize_output_size_scale_below_upper_bound,
+)
+
+
+def get_capped_longest_edge(height: int, width: int, longest_edge: int, tile_size: int, max_image_tiles: int) -> int:
+    for edge in range(longest_edge, tile_size - 1, -tile_size):
+        resized_height, resized_width = _resize_output_size_rescale_to_max_len(height, width, max_len=edge)
+        resized_height, resized_width = _resize_output_size_scale_below_upper_bound(
+            resized_height, resized_width, max_len=MAX_IMAGE_SIZE
+        )
+        aspect_ratio = resized_width / resized_height
+        if resized_width >= resized_height:
+            cols = math.ceil(resized_width / tile_size)
+            rows = math.ceil(int(cols * tile_size / aspect_ratio) / tile_size)
+        else:
+            rows = math.ceil(resized_height / tile_size)
+            cols = math.ceil(int(rows * tile_size * aspect_ratio) / tile_size)
+        if rows * cols <= max_image_tiles:
+            return edge
+    return tile_size
+
+
+class UltraVBertImageProcessorKwargs(Idefics3ImageProcessorKwargs, total=False):
+    max_image_tiles: int | None
+
+
+class UltraVBertImageProcessor(Idefics3ImageProcessor):
+    max_image_tiles = None
+    valid_kwargs = UltraVBertImageProcessorKwargs
+
+    def _preprocess(self, images, *, do_resize, size, resample, do_image_splitting, max_image_size,
+                    max_image_tiles=None, **kwargs):
+        if do_resize and do_image_splitting and max_image_tiles is not None:
+            tile_size = max_image_size["longest_edge"]
+            images = [
+                [
+                    self.resize(
+                        image,
+                        SizeDict(
+                            longest_edge=get_capped_longest_edge(
+                                *image.shape[-2:], size.longest_edge, tile_size, max_image_tiles
+                            )
+                        ),
+                        resample=resample,
+                    )
+                    for image in batch_images
+                ]
+                for batch_images in images
+            ]
+            do_resize = False
+        return super()._preprocess(images, do_resize=do_resize, size=size, resample=resample,
+                                   do_image_splitting=do_image_splitting, max_image_size=max_image_size, **kwargs)
+
+    def get_number_of_image_patches(self, height: int, width: int, images_kwargs: dict | None = None):
+        images_kwargs = dict(images_kwargs or {})
+        max_image_tiles = images_kwargs.get("max_image_tiles", self.max_image_tiles)
+        if max_image_tiles is not None:
+            size = images_kwargs.get("size", self.size)
+            tile_size = images_kwargs.get("max_image_size", self.max_image_size)["longest_edge"]
+            longest_edge = get_capped_longest_edge(height, width, size["longest_edge"], tile_size, max_image_tiles)
+            images_kwargs["size"] = {"longest_edge": longest_edge}
+        return super().get_number_of_image_patches(height, width, images_kwargs)
+
+
+__all__ = ["UltraVBertImageProcessor"]

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:29512ca01cbdae4b0ba9e45d7a36a30dc6a28506ecba8a583afa2b9cf53ad3af
+size 586285080

modeling_ultravbert.py

@@ -0,0 +1,126 @@
+from torch import nn
+from torch.nn import CrossEntropyLoss
+from transformers import AutoModel
+from transformers.models.modernvbert.modeling_modernvbert import (
+    ModernVBertConnector,
+    ModernVBertForSequenceClassification,
+    ModernVBertForTokenClassification,
+    ModernVBertMaskedLMOutput,
+    ModernVBertModel,
+    ModernVBertPreTrainedModel,
+)
+from transformers.models.ultrabert.modeling_ultrabert import UltrabertPredictionHead, UltrabertRMSNorm
+
+from .configuration_ultravbert import UltraVBertConfig
+
+
+class UltraVBertConnector(ModernVBertConnector):
+    def __init__(self, config: UltraVBertConfig):
+        super().__init__(config)
+        # UltraBERT's first layer takes embeddings unnormalized; raw projections would swamp its token embeddings.
+        self.norm = (
+            UltrabertRMSNorm(config.text_config.hidden_size, eps=config.text_config.rms_norm_eps)
+            if config.connector_norm
+            else nn.Identity()
+        )
+
+    def forward(self, image_hidden_states):
+        return self.norm(super().forward(image_hidden_states))
+
+
+class UltraVBertModel(ModernVBertModel):
+    config_class = UltraVBertConfig
+    _keys_to_ignore_on_load_unexpected = [r"vision_model\.head\."]
+
+    def __init__(self, config: UltraVBertConfig):
+        # ModernVBertModel.__init__ builds ModernVBertConnector; the rest of its body follows unchanged.
+        ModernVBertPreTrainedModel.__init__(self, config)
+        self.padding_idx = config.text_config.pad_token_id
+        self.vocab_size = config.text_config.vocab_size
+        self.vision_model = AutoModel.from_config(config.vision_config)
+        self.connector = UltraVBertConnector(config)
+        self.text_model = AutoModel.from_config(config.text_config)
+        self.image_seq_len = int(
+            ((config.vision_config.image_size // config.vision_config.patch_size) ** 2)
+            / (config.pixel_shuffle_factor**2)
+        )
+        self.image_token_id = config.image_token_id
+        self.post_init()
+
+
+class UltraVBertForMaskedLM(ModernVBertPreTrainedModel):
+    config_class = UltraVBertConfig
+    _keys_to_ignore_on_load_unexpected = [r"vision_model\.head\."]
+    _tied_weights_keys = {"head.decoder.weight": "model.text_model.embed_tokens.weight"}
+
+    def __init__(self, config: UltraVBertConfig):
+        super().__init__(config)
+        self.vocab_size = config.text_config.vocab_size
+        self.model = UltraVBertModel(config)
+        self.head = UltrabertPredictionHead(config.text_config)
+        self.post_init()
+
+    def get_output_embeddings(self):
+        return self.head.decoder
+
+    def set_output_embeddings(self, new_embeddings):
+        self.head.decoder = new_embeddings
+
+    def forward(self, input_ids=None, attention_mask=None, position_ids=None, inputs_embeds=None, pixel_values=None,
+                pixel_attention_mask=None, image_hidden_states=None, labels=None, **kwargs):
+        outputs = self.model(
+            input_ids=input_ids,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            inputs_embeds=inputs_embeds,
+            pixel_values=pixel_values,
+            pixel_attention_mask=pixel_attention_mask,
+            image_hidden_states=image_hidden_states,
+            **kwargs,
+        )
+        logits = self.head(outputs.last_hidden_state)
+        loss = None
+        if labels is not None:
+            loss = CrossEntropyLoss()(logits.view(-1, self.vocab_size), labels.view(-1))
+        return ModernVBertMaskedLMOutput(
+            loss=loss,
+            logits=logits,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+            image_hidden_states=outputs.image_hidden_states,
+        )
+
+
+def _build_classifier(model, config: UltraVBertConfig):
+    # ModernVBert's own __init__ builds a ModernBERT prediction head that UltraBERT's text config cannot configure.
+    ModernVBertPreTrainedModel.__init__(model, config)
+    model.num_labels = config.num_labels
+    model.model = UltraVBertModel(config)
+    model.head = nn.Identity()
+    model.drop = nn.Dropout(config.classifier_dropout)
+    model.classifier = nn.Linear(config.text_config.hidden_size, config.num_labels)
+    model.post_init()
+
+
+class UltraVBertForSequenceClassification(ModernVBertForSequenceClassification):
+    config_class = UltraVBertConfig
+    _keys_to_ignore_on_load_unexpected = [r"vision_model\.head\."]
+
+    def __init__(self, config: UltraVBertConfig):
+        _build_classifier(self, config)
+
+
+class UltraVBertForTokenClassification(ModernVBertForTokenClassification):
+    config_class = UltraVBertConfig
+    _keys_to_ignore_on_load_unexpected = [r"vision_model\.head\."]
+
+    def __init__(self, config: UltraVBertConfig):
+        _build_classifier(self, config)
+
+
+__all__ = [
+    "UltraVBertModel",
+    "UltraVBertForMaskedLM",
+    "UltraVBertForSequenceClassification",
+    "UltraVBertForTokenClassification",
+]

processor_config.json

@@ -0,0 +1,35 @@
+{
+  "image_processor": {
+    "do_convert_rgb": true,
+    "do_image_splitting": true,
+    "do_normalize": true,
+    "do_pad": true,
+    "do_rescale": true,
+    "do_resize": true,
+    "image_mean": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "image_processor_type": "UltraVBertImageProcessor",
+    "image_std": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "max_image_size": {
+      "longest_edge": 512
+    },
+    "max_image_tiles": 27,
+    "resample": 1,
+    "rescale_factor": 0.00392156862745098,
+    "size": {
+      "longest_edge": 3072
+    },
+    "auto_map": {
+      "AutoImageProcessor": "image_processing_ultravbert.UltraVBertImageProcessor"
+    }
+  },
+  "image_seq_len": 64,
+  "processor_class": "Idefics3Processor"
+}

tokenizer_config.json

@@ -0,0 +1,23 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<fake_token_around_image>",
+    "<image>",
+    "<end_of_utterance>"
+  ],
+  "is_local": true,
+  "local_files_only": false,
+  "mask_token": "<|mask|>",
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "processor_class": "Idefics3Processor",
+  "sep_token": "<|sep|>",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>",
+  "vocab_size": 49152
+}