decosa-oneline-reader-paddleocr-vl v1: PaddleOCR-VL-1.6 fine-tune for one-line diagrams (Apache-2.0)

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* 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
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

LICENSE

@@ -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 (c) 2025 PaddlePaddle Authors. All Rights Reserved.
+
+   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.

NOTICE

@@ -0,0 +1,12 @@
+decosa-oneline-reader-paddleocr-vl
+Copyright 2026 Decosa. Licensed under the Apache License, Version 2.0 (see LICENSE).
+
+Base model: PaddleOCR-VL-1.6 (PaddlePaddle/PaddleOCR-VL-1.6, revision c5630abae1d940eafe0697512a0325494b02ab42),
+Copyright (c) 2025 PaddlePaddle Authors, Apache License 2.0. This repository contains a full fine-tune of its weights.
+The files configuration_paddleocr_vl.py, modeling_paddleocr_vl.py, image_processing_paddleocr_vl.py and
+processing_paddleocr_vl.py are copied unchanged from the base repository (Apache License 2.0, PaddlePaddle Authors).
+The tokenizer and chat template come from the base model.
+
+Training data: synthetic one-line diagrams drawn by a Decosa script (fictional plants, places and equipment). Text was
+rendered with the Liberation fonts (SIL Open Font License 1.1, Red Hat / Google) and DejaVu fonts (Bitstream Vera and
+DejaVu licence). No font files, drawings or training data are redistributed here.

README.md

@@ -0,0 +1,127 @@
+---
+license: apache-2.0
+language: [en]
+pipeline_tag: image-text-to-text
+library_name: transformers
+base_model: PaddlePaddle/PaddleOCR-VL-1.6
+tags: [document-ai, ocr, engineering-drawings, one-line-diagram, single-line-diagram, interconnection, energy]
+---
+
+# decosa-oneline-reader-paddleocr-vl
+
+A 0.96B vision-language model that reads an electrical one-line (single-line) diagram for a solar or storage
+interconnection request, including poor fax-like scans, and returns one JSON object: the point-of-interconnection
+voltage, the plant output limit, every inverter group (quantity, model, rating), battery storage, the main transformer,
+and whether a disconnect switch, meter and main breaker are drawn.
+
+It is a full fine-tune of [PaddleOCR-VL-1.6](https://huggingface.co/PaddlePaddle/PaddleOCR-VL-1.6). It was trained for
+one job: turning the drawing in an interconnection packet into fields that can be checked against the application form.
+It is not a general OCR or document parser any more.
+
+## Results
+
+Four test sets of 60 drawings each, drawn by the same procedural generator as the training data but with seeds that
+were never used for training. "All fields right" means every field of the drawing is right (9 fields: POI kV, plant
+limit, inverter models, counts and kVA, storage MW and MWh, transformer MVA, disconnect). The baselines read the same
+images with the same scorer.
+
+| Test set (60 drawings each) | Base PaddleOCR-VL-1.6 | Qwen3.8-27B (vision) | **This model** |
+|---|---|---|---|
+| Poor scans (scaled down, blurred, skewed, speckled, JPEG) | 0 of 12 tried (no usable JSON) | 25 | **59** |
+| The clean set below, degraded the same way | — | 16 | **59** |
+| Clean drawings (regression check) | 0 of 12 tried | 59 | **60** |
+| Poor scans where every inverter model name and rating is random (never seen in training) | — | — | **50** |
+
+Per field on the poor scans, this model vs Qwen3.8-27B: inverter models 60 vs 38 of 60, inverter counts 91 vs 57 of 91,
+inverter kVA 91 vs 67 of 91; every other field 98-100% for both. On the random-name set the misses are one wrong
+character in a model name (for example 8908 read as 6908), a hyphen read as a space, or a count off by one.
+
+Speed: about 1.2 s per drawing in bf16 on one workstation GPU, batch 6 (2-3 s on the larger clean drawings).
+
+All numbers are in `eval_summary.json`.
+
+## Intended use
+
+- First-pass reading of one-line diagrams in interconnection packets, so a reviewer can compare the drawing with the
+  application form (capacity, inverter count, POI voltage, storage) without retyping it.
+- A second reader next to a general vision model on scans that model struggles with.
+
+Not for: approving or rejecting an interconnection request, engineering or protection studies, or any decision without
+a person checking the drawing. Treat every value as a reading to verify, not a fact.
+
+## Limitations (read these)
+
+- **One drawing family.** Training and every test set come from one procedural drawing template (four label styles,
+  three font sets, many degradations). The results show it reads that family well, even when badly scanned. **No real
+  utility one-line has been scored.** Real drawings (CAD title blocks, dense protection schemes, several sheets,
+  hand mark-ups) are out of distribution; expect missed or invented fields until it is tested and re-trained on them.
+- Invented model names: about 1 drawing in 8 has a one-character slip in a model name. Check model strings against the
+  application or a datasheet list.
+- It reports what it read; it does not know whether a value is plausible. Pair it with arithmetic checks (inverter
+  count × rating vs plant limit) and a person.
+- English labels, US-style units (kV, kVA, MW, MWh) only.
+
+## Training
+
+| | |
+|---|---|
+| Base | PaddlePaddle/PaddleOCR-VL-1.6 @ c5630abae1d9 (ERNIE-4.5-0.3B language model + NaViT vision encoder), Apache-2.0 |
+| Data | 16,000 synthetic one-line diagrams drawn by a Decosa script: fictional plants, places and inverter models; 35% clean, 15% one fixed poor-scan degradation, 50% random degradations (scale 0.42-0.9, blur, skew up to 2.5°, fax threshold, grey paper, JPEG 20-75); half the drawings use random fictional model names and ratings. Fonts: Liberation (SIL OFL 1.1) and DejaVu. No real drawings, no personal data, nothing scraped. The generator and data are not released. |
+| Target | The compact JSON below, one object per drawing; every training target was parsed back and checked against the drawing's ground truth before use |
+| Method | Full fine-tune, AdamW (β 0.9/0.95), lr 3e-5 cosine with 40 warm-up steps, vision tower at 0.3× lr, bf16 autocast with fp32 master weights, loss on answer tokens only, token-budgeted batches (24k tokens), 1,707 steps (58 min on one H200), seed 20260928. Validation loss 2.55 → 0.0041. |
+| Selection | No choice was made on the test sets; the final checkpoint is the end of the time-boxed run. |
+
+## Usage
+
+```bash
+pip install "transformers>=5.17" torch pillow
+python usage.py drawing.png
+```
+
+```python
+import json, torch
+from PIL import Image
+from transformers import AutoModelForImageTextToText, AutoProcessor
+
+repo = "decosaai/decosa-oneline-reader-paddleocr-vl"
+proc = AutoProcessor.from_pretrained(repo)
+model = AutoModelForImageTextToText.from_pretrained(repo, dtype=torch.bfloat16).to("cuda").eval()
+
+im = Image.open("drawing.png").convert("RGB")      # see usage.py: small scans are enlarged to ~1 MP first
+msgs = [{"role": "user", "content": [{"type": "image", "image": im},
+                                     {"type": "text", "text": "One-line Diagram Recognition:"}]}]
+x = proc.apply_chat_template(msgs, add_generation_prompt=True, tokenize=True, return_dict=True,
+                             return_tensors="pt").to("cuda")
+out = model.generate(**x, max_new_tokens=900, do_sample=False, use_cache=True)
+reading = json.loads(proc.decode(out[0][x["input_ids"].shape[-1]:], skip_special_tokens=True))
+```
+
+Pass `use_cache=True` to `generate()`: the base config ships with the cache off in the text config, which makes
+generation many times slower. Tested with transformers 5.17, which loads the model natively (no `trust_remote_code`);
+the base repo's code files are included only because `config.json` still names them in `auto_map`.
+
+Output (field order as trained; values here are placeholders):
+
+```json
+{"is_one_line": true, "poi_kv": 0, "poi_label": "...",
+ "plant_limit": {"value": 0, "unit": "MW", "text": "..."},
+ "inverter_groups": [{"text": "...", "qty": 0, "model": "...", "rating": 0, "rating_unit": "kVA", "type": "pv"}],
+ "total_inverter_note": {"value": null, "unit": "kVA", "text": ""},
+ "storage": {"mw": 0, "mwh": 0, "text": "..."},
+ "transformer": {"text": "...", "max_rating": 0, "rating_unit": "MVA", "hv_kv": 0, "lv_kv": 0},
+ "disconnect": {"present": true, "lockable": true, "text": "..."},
+ "meter": {"present": true, "text": "..."},
+ "main_breaker": {"present": true, "text": "..."}}
+```
+
+One `inverter_groups` entry per feeder branch, even when the same model repeats; quantities are copied as drawn, not
+summed.
+
+Files: `model.safetensors` (bf16), `config.json`, `generation_config.json`, `processor_config.json`, tokenizer files,
+`chat_template.jinja`, the base model's `*_paddleocr_vl.py` code files (unchanged, Apache-2.0, PaddlePaddle Authors),
+`usage.py`, `eval_summary.json`, `SHA256SUMS`, `LICENSE`, `NOTICE`.
+
+## Licence
+
+Apache-2.0 for the weights and `usage.py`. The base model and its code files are Apache-2.0, Copyright PaddlePaddle
+Authors; see NOTICE.

SHA256SUMS

@@ -0,0 +1,16 @@
+2f27812dab7f333e471884e0c803d807f11953d5453140dfb1aaba234f872bc8  chat_template.jinja
+9c0379f6cc68c8ad950bd76716c40e3cbb4e502f96fe7dcc591e147173c92262  config.json
+753dd93654c3a9c8c85a3eaee1e3092dd12591b0f2dce0305e1abfb7a41ff160  configuration_paddleocr_vl.py
+7f0b73aa1bbffa1f9572ca7499aaec6bfc96f6b5beef680a472b4f4a00d0adc8  eval_summary.json
+d34a3c0ceac7541c8d89c3eafb685adf5e3f5c63f27fbe8f7bb03bf1cfe1fa6e  generation_config.json
+a4fa521b9cb16e207f94b7f2d16427771776dfc634420d319fc4916ee58049ec  image_processing_paddleocr_vl.py
+b8c4d7deccd236af023af1c88c4d4e8f0fc2f41914e0fb23a3ec9678fb5a8456  LICENSE
+c5013dff57ca8b87dc1de64d0fd839a44313de09d230a4fb2d08289d2cad5111  modeling_paddleocr_vl.py
+d1b2613b26009260e1f98cf691f14da9f9051ad3880a172cc428d1f7f05934e7  model.safetensors
+cafb13672ad15bcadb55b77f76a12b73c5842fd6f985e6d6e9525ab9a06d9041  NOTICE
+e29cb1e5f275f2bd3ce051bd5c9983a33894e693b2823a0e13d4c07c8c4f9e13  processing_paddleocr_vl.py
+fcf182714867cf2a401d644037a8b24678ff0780da5e14783bcc1f1b1c0ed21e  processor_config.json
+8361f04a4e022dd4fa232b796fa40ccb7c81627df7c9d06c130b5b9dfe7a0b7f  README.md
+0daa4d084c1b7a7343d1c5fbad2e933ca1f63c6858d8db0d72955e9e0bb93356  tokenizer_config.json
+c8a215a59183d0d0781adc33bacd3ce6162716f7fd568fb30234a74d69803a7d  tokenizer.json
+f5ad20e35f9a323aacc3a76409f346770e96aa3cd9f7cb702066fd38e436e5f9  usage.py

chat_template.jinja

@@ -0,0 +1,43 @@
+{%- if not add_generation_prompt is defined -%}
+    {%- set add_generation_prompt = true -%}
+{%- endif -%}
+{%- if not cls_token is defined -%}
+    {%- set cls_token = "<|begin_of_sentence|>" -%}
+{%- endif -%}
+{%- if not eos_token is defined -%}
+    {%- set eos_token = "</s>" -%}
+{%- endif -%}
+{{- cls_token -}}
+{%- for message in messages -%}
+    {%- if message["role"] == "user" -%}
+        {{- "User: " -}}
+        {%- for content in message["content"] -%}
+            {%- if content["type"] == "image" -%}
+                {{ "<|IMAGE_START|><|IMAGE_PLACEHOLDER|><|IMAGE_END|>" }}
+            {%- endif -%}
+        {%- endfor -%}
+        {%- for content in message["content"] -%}
+            {%- if content["type"] == "text" -%}
+                {{ content["text"] }}
+            {%- endif -%}
+        {%- endfor -%}
+        {{ "\n" -}}
+    {%- elif message["role"] == "assistant" -%}
+        {{- "Assistant:\n" -}}
+        {%- for content in message["content"] -%}
+            {%- if content["type"] == "text" -%}
+                {{ content["text"] }}
+            {%- endif -%}
+        {%- endfor -%}
+        {{ eos_token -}}
+    {%- elif message["role"] == "system" -%}
+        {%- for content in message["content"] -%}
+            {%- if content["type"] == "text" -%}
+                {{ content["text"] + "\n" }}
+            {%- endif -%}
+        {%- endfor -%}
+    {%- endif -%}
+{%- endfor -%}
+{%- if add_generation_prompt -%}
+    {{- "Assistant:\n" -}}
+{%- endif -%}

config.json

@@ -0,0 +1,89 @@
+{
+  "architectures": [
+    "PaddleOCRVLForConditionalGeneration"
+  ],
+  "attention_probs_dropout_prob": 0.0,
+  "auto_map": {
+    "AutoConfig": "configuration_paddleocr_vl.PaddleOCRVLConfig",
+    "AutoModel": "modeling_paddleocr_vl.PaddleOCRVLForConditionalGeneration",
+    "AutoModelForCausalLM": "modeling_paddleocr_vl.PaddleOCRVLForConditionalGeneration"
+  },
+  "compression_ratio": 1.0,
+  "dtype": "float32",
+  "hidden_dropout_prob": 0.0,
+  "ignored_index": -100,
+  "image_token_id": 100295,
+  "max_sequence_length": null,
+  "model_type": "paddleocr_vl",
+  "rope_is_neox_style": true,
+  "sliding_window": null,
+  "text_config": {
+    "bos_token_id": 1,
+    "dtype": "float32",
+    "eos_token_id": 2,
+    "head_dim": 128,
+    "hidden_act": "silu",
+    "hidden_size": 1024,
+    "initializer_range": 0.02,
+    "intermediate_size": 3072,
+    "max_position_embeddings": 131072,
+    "model_type": "paddleocr_vl_text",
+    "num_attention_heads": 16,
+    "num_hidden_layers": 18,
+    "num_key_value_heads": 2,
+    "pad_token_id": 0,
+    "rms_norm_eps": 1e-05,
+    "rope_parameters": {
+      "mrope_section": [
+        16,
+        24,
+        24
+      ],
+      "rope_theta": 500000,
+      "rope_type": "default",
+      "type": "default"
+    },
+    "tie_word_embeddings": true,
+    "use_bias": false,
+    "use_cache": false,
+    "vocab_size": 103424
+  },
+  "tie_word_embeddings": true,
+  "transformers_version": "5.17.0",
+  "use_3d_rope": true,
+  "use_cache": true,
+  "use_flash_attention": false,
+  "video_token_id": 101307,
+  "vision_config": {
+    "architectures": [
+      "PaddleOCRVisionModel"
+    ],
+    "attention_dropout": 0.0,
+    "auto_map": {
+      "AutoConfig": "configuration_paddleocr_vl.PaddleOCRVLConfig",
+      "AutoModel": "modeling_paddleocr_vl.PaddleOCRVisionModel"
+    },
+    "dtype": "float32",
+    "hidden_act": "gelu_pytorch_tanh",
+    "hidden_size": 1152,
+    "image_size": 384,
+    "intermediate_size": 4304,
+    "layer_norm_eps": 1e-06,
+    "model_type": "paddleocr_vl_vision",
+    "num_attention_heads": 16,
+    "num_channels": 3,
+    "num_hidden_layers": 27,
+    "pad_token_id": 0,
+    "patch_size": 14,
+    "rope_parameters": {
+      "rope_theta": 10000.0,
+      "rope_type": "axial"
+    },
+    "spatial_merge_size": 2,
+    "temporal_patch_size": 2,
+    "tokens_per_second": 2
+  },
+  "vision_end_token_id": 101306,
+  "vision_start_token_id": 101305,
+  "weight_share_add_bias": true
+}

configuration_paddleocr_vl.py

@@ -0,0 +1,191 @@
+# Copyright (c) 2025 PaddlePaddle Authors. All Rights Reserved.
+#
+# 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.
+
+from transformers.configuration_utils import PretrainedConfig
+from transformers.modeling_rope_utils import rope_config_validation
+
+class PaddleOCRVisionConfig(PretrainedConfig):
+    model_type = "paddleocr_vl"
+    base_config_key = "vision_config"
+
+    def __init__(
+        self,
+        hidden_size=768,
+        intermediate_size=3072,
+        num_hidden_layers=12,
+        num_attention_heads=12,
+        num_channels=3,
+        image_size=224,
+        patch_size=14,
+        hidden_act="gelu_pytorch_tanh",
+        layer_norm_eps=1e-6,
+        attention_dropout=0.0,
+        spatial_merge_size=2,
+        temporal_patch_size=2,
+        tokens_per_second=2,
+        **kwargs,
+    ):
+        super().__init__(**kwargs)
+
+        self.hidden_size = hidden_size
+        self.intermediate_size = intermediate_size
+        self.num_hidden_layers = num_hidden_layers
+        self.num_attention_heads = num_attention_heads
+        self.num_channels = num_channels
+        self.patch_size = patch_size
+        self.image_size = image_size
+        self.attention_dropout = attention_dropout
+        self.layer_norm_eps = layer_norm_eps
+        self.hidden_act = hidden_act
+        self.spatial_merge_size = spatial_merge_size
+        self.temporal_patch_size = temporal_patch_size
+        self.tokens_per_second = tokens_per_second
+
+
+
+class PaddleOCRVLConfig(PretrainedConfig):
+    """
+    Configuration class.
+
+    This class stores the configuration of an Ernie model, defining the model architecture.
+    It inherits from PretrainedConfig and can be used to control model outputs.
+    """
+
+    model_type = "paddleocr_vl"
+    keys_to_ignore_at_inference = ["past_key_values"]
+    sub_configs = {"vision_config": PaddleOCRVisionConfig}
+
+    # Default tensor parallel plan for base model `Qwen3`
+    base_model_tp_plan = {
+        "layers.*.self_attn.q_proj": "colwise",
+        "layers.*.self_attn.k_proj": "colwise",
+        "layers.*.self_attn.v_proj": "colwise",
+        "layers.*.self_attn.o_proj": "rowwise",
+        "layers.*.mlp.gate_proj": "colwise",
+        "layers.*.mlp.up_proj": "colwise",
+        "layers.*.mlp.down_proj": "rowwise",
+    }
+    base_model_pp_plan = {
+        "embed_tokens": (["input_ids"], ["inputs_embeds"]),
+        "layers": (["hidden_states", "attention_mask"], ["hidden_states"]),
+        "norm": (["hidden_states"], ["hidden_states"]),
+    }
+
+    def __init__(
+        self,
+        vocab_size=32000,
+        hidden_size=768,
+        intermediate_size=11008,
+        max_position_embeddings=32768,
+        num_hidden_layers=2,
+        num_attention_heads=2,
+        image_token_id=101304,
+        video_token_id=101305,
+        vision_start_token_id=101306,
+        rms_norm_eps=1e-6,
+        use_cache=False,
+        use_flash_attention=False,
+        pad_token_id=0,
+        bos_token_id=1,
+        eos_token_id=2,
+        head_dim=128,
+        hidden_act="silu",
+        use_bias=False,
+        rope_theta=10000,
+        weight_share_add_bias=True,
+        ignored_index=-100,
+        attention_probs_dropout_prob=0.0,
+        hidden_dropout_prob=0.0,
+        compression_ratio: float = 1.0,
+        num_key_value_heads=None,
+        max_sequence_length=None,
+        tie_word_embeddings=False,
+        vision_config=None,
+        rope_scaling=None,
+        **kwargs,
+    ):
+        """
+        Initialize configuration with default or specified parameters.
+
+        Args:
+            vocab_size (int): Size of the vocabulary (number of unique tokens)
+            hidden_size (int): Dimensionality of the encoder layers and the pooler layer
+            intermediate_size (int): Dimensionality of the "intermediate" (feed-forward) layer
+            max_position_embeddings (int): Maximum sequence length the model can handle
+            num_hidden_layers (int): Number of hidden layers in the Transformer encoder
+            num_attention_heads (int): Number of attention heads for each attention layer
+            rms_norm_eps (float): The epsilon used by the RMS normalization layers
+            use_cache (bool): Whether to use caching for faster generation (decoding)
+            use_flash_attention (bool): Whether to use FlashAttention for optimized attention computation
+            pad_token_id (int): Token ID used for padding sequences
+            bos_token_id (int): Token ID used for beginning-of-sequence
+            eos_token_id (int): Token ID used for end-of-sequence
+            use_bias (bool): Whether to use bias terms in linear layers
+            rope_theta (float): The base period of the RoPE embeddings
+            weight_share_add_bias (bool): Whether to share bias weights in certain layers
+            ignored_index (int): Target value that is ignored during loss computation
+            attention_probs_dropout_prob (float): Dropout probability for attention weights
+            hidden_dropout_prob (float): Dropout probability for hidden layers
+            compression_ratio (float): Ratio for KV cache compression (1.0 = no compression)
+            num_key_value_heads (int): Number of key/value heads (for Grouped Query Attention)
+            max_sequence_length (int): Maximum sequence length for positional embeddings
+            **kwargs: Additional keyword arguments passed to parent class
+        """
+
+        # Set default for tied embeddings if not specified.
+        super().__init__(
+            pad_token_id=pad_token_id,
+            bos_token_id=bos_token_id,
+            eos_token_id=eos_token_id,
+            **kwargs,
+        )
+        if isinstance(vision_config, dict):
+            self.vision_config = self.sub_configs["vision_config"](**vision_config)
+        elif vision_config is None:
+            self.vision_config = self.sub_configs["vision_config"]()        
+        self.vocab_size = vocab_size
+        self.hidden_size = hidden_size
+        self.intermediate_size = intermediate_size
+        self.max_position_embeddings = max_position_embeddings
+        self.num_hidden_layers = num_hidden_layers
+        self.num_attention_heads = num_attention_heads
+        self.rms_norm_eps = rms_norm_eps
+        self.use_cache = use_cache
+        self.use_flash_attention = use_flash_attention
+        self.pad_token_id = pad_token_id
+        self.bos_token_id = bos_token_id
+        self.eos_token_id = eos_token_id
+        self.image_token_id = image_token_id
+        self.video_token_id = video_token_id
+        self.vision_start_token_id = vision_start_token_id
+        self.head_dim = head_dim
+        self.hidden_act=hidden_act
+        self.sliding_window = None
+        self.hidden_size = hidden_size
+        self.use_bias = use_bias
+        self.weight_share_add_bias = weight_share_add_bias
+        self.rope_theta = rope_theta
+        self.ignored_index = ignored_index
+        self.attention_probs_dropout_prob = attention_probs_dropout_prob
+        self.hidden_dropout_prob = hidden_dropout_prob
+        self.compression_ratio = compression_ratio
+        self.num_key_value_heads = num_key_value_heads
+        self.max_sequence_length = max_sequence_length
+        self.rope_scaling = rope_scaling
+        if self.rope_scaling is not None and "type" in self.rope_scaling:
+            if self.rope_scaling["type"] == "mrope":
+                self.rope_scaling["type"] = "default"
+            self.rope_scaling["rope_type"] = self.rope_scaling["type"]
+        rope_config_validation(self, ignore_keys={"mrope_section"})        
+        super().__init__(tie_word_embeddings=tie_word_embeddings, **kwargs)

eval_summary.json

@@ -0,0 +1,330 @@
+{
+ "model": "decosa-oneline-reader-paddleocr-vl",
+ "date": "2026-09-28",
+ "base": "PaddlePaddle/PaddleOCR-VL-1.6@c5630abae1d940eafe0697512a0325494b02ab42",
+ "metric": "all_fields_right per drawing (9 fields: poi_kv, plant_limit_mw, models, qty, kva, bess_mw, bess_mwh, transformer_mva, has_disconnect); per-field right/n",
+ "setup": "test sets of 60 synthetic drawings each from the training generator family, seeds disjoint from training; bf16, greedy, max_new_tokens 900, batch 6; a re-run on a second machine gave identical results",
+ "caveat": "no real utility one-line diagram has been scored",
+ "sets": {
+  "hard2": {
+   "desc": "poor scans (fresh), the headline set",
+   "this_model": {
+    "all_fields_right": 59,
+    "cases": 60,
+    "fields": {
+     "poi_kv": [
+      60,
+      60
+     ],
+     "plant_limit_mw": [
+      60,
+      60
+     ],
+     "models": [
+      60,
+      60
+     ],
+     "qty": [
+      91,
+      91
+     ],
+     "kva": [
+      91,
+      91
+     ],
+     "bess_mw": [
+      49,
+      49
+     ],
+     "bess_mwh": [
+      48,
+      49
+     ],
+     "transformer_mva": [
+      60,
+      60
+     ],
+     "has_disconnect": [
+      60,
+      60
+     ]
+    }
+   },
+   "seconds_total": 69.3,
+   "qwen3.8-27b_vision": {
+    "all_fields_right": 25,
+    "cases": 60,
+    "fields": {
+     "poi_kv": [
+      60,
+      60
+     ],
+     "plant_limit_mw": [
+      59,
+      60
+     ],
+     "models": [
+      38,
+      60
+     ],
+     "qty": [
+      57,
+      91
+     ],
+     "kva": [
+      67,
+      91
+     ],
+     "bess_mw": [
+      48,
+      49
+     ],
+     "bess_mwh": [
+      48,
+      49
+     ],
+     "transformer_mva": [
+      56,
+      60
+     ],
+     "has_disconnect": [
+      60,
+      60
+     ]
+    }
+   },
+   "base_first12": {
+    "all_fields_right": 0,
+    "cases": 12,
+    "parsed": 0
+   }
+  },
+  "hard": {
+   "desc": "the clean set degraded",
+   "this_model": {
+    "all_fields_right": 59,
+    "cases": 60,
+    "fields": {
+     "poi_kv": [
+      60,
+      60
+     ],
+     "plant_limit_mw": [
+      60,
+      60
+     ],
+     "models": [
+      60,
+      60
+     ],
+     "qty": [
+      88,
+      88
+     ],
+     "kva": [
+      88,
+      88
+     ],
+     "bess_mw": [
+      48,
+      48
+     ],
+     "bess_mwh": [
+      47,
+      48
+     ],
+     "transformer_mva": [
+      60,
+      60
+     ],
+     "has_disconnect": [
+      60,
+      60
+     ]
+    }
+   },
+   "seconds_total": 137.8,
+   "qwen3.8-27b_vision": {
+    "all_fields_right": 16,
+    "cases": 60,
+    "fields": {
+     "poi_kv": [
+      60,
+      60
+     ],
+     "plant_limit_mw": [
+      53,
+      60
+     ],
+     "models": [
+      34,
+      60
+     ],
+     "qty": [
+      43,
+      88
+     ],
+     "kva": [
+      57,
+      88
+     ],
+     "bess_mw": [
+      34,
+      48
+     ],
+     "bess_mwh": [
+      35,
+      48
+     ],
+     "transformer_mva": [
+      46,
+      60
+     ],
+     "has_disconnect": [
+      60,
+      60
+     ]
+    }
+   }
+  },
+  "crisp": {
+   "desc": "clean drawings (regression check)",
+   "this_model": {
+    "all_fields_right": 60,
+    "cases": 60,
+    "fields": {
+     "poi_kv": [
+      60,
+      60
+     ],
+     "plant_limit_mw": [
+      60,
+      60
+     ],
+     "models": [
+      60,
+      60
+     ],
+     "qty": [
+      88,
+      88
+     ],
+     "kva": [
+      88,
+      88
+     ],
+     "bess_mw": [
+      48,
+      48
+     ],
+     "bess_mwh": [
+      48,
+      48
+     ],
+     "transformer_mva": [
+      60,
+      60
+     ],
+     "has_disconnect": [
+      60,
+      60
+     ]
+    }
+   },
+   "seconds_total": 120.3,
+   "qwen3.8-27b_vision": {
+    "all_fields_right": 59,
+    "cases": 60,
+    "fields": {
+     "poi_kv": [
+      60,
+      60
+     ],
+     "plant_limit_mw": [
+      60,
+      60
+     ],
+     "models": [
+      60,
+      60
+     ],
+     "qty": [
+      87,
+      88
+     ],
+     "kva": [
+      88,
+      88
+     ],
+     "bess_mw": [
+      48,
+      48
+     ],
+     "bess_mwh": [
+      48,
+      48
+     ],
+     "transformer_mva": [
+      60,
+      60
+     ],
+     "has_disconnect": [
+      60,
+      60
+     ]
+    }
+   },
+   "base_first12": {
+    "all_fields_right": 0,
+    "cases": 12,
+    "parsed": 0
+   }
+  },
+  "novel": {
+   "desc": "poor scans with random inverter model names and ratings never used in training",
+   "this_model": {
+    "all_fields_right": 50,
+    "cases": 60,
+    "fields": {
+     "poi_kv": [
+      60,
+      60
+     ],
+     "plant_limit_mw": [
+      60,
+      60
+     ],
+     "models": [
+      53,
+      60
+     ],
+     "qty": [
+      89,
+      98
+     ],
+     "kva": [
+      91,
+      98
+     ],
+     "bess_mw": [
+      46,
+      47
+     ],
+     "bess_mwh": [
+      47,
+      47
+     ],
+     "transformer_mva": [
+      60,
+      60
+     ],
+     "has_disconnect": [
+      60,
+      60
+     ]
+    }
+   },
+   "seconds_total": 105.8
+  }
+ }
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "eos_token_id": 2,
+  "pad_token_id": 0,
+  "transformers_version": "5.17.0",
+  "use_cache": true
+}

image_processing_paddleocr_vl.py

@@ -0,0 +1,570 @@
+# Copyright (c) 2025 PaddlePaddle Authors. All Rights Reserved.
+#
+# 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.
+
+"""Image processor class for PaddleOCR-VL."""
+
+import math
+from typing import Dict, List, Optional, Union
+
+import numpy as np
+import torch
+from transformers.image_processing_utils import BaseImageProcessor, BatchFeature
+from torchvision.transforms import functional as TF
+from transformers.image_transforms import (
+    convert_to_rgb,
+    resize,
+    to_channel_dimension_format,
+)
+from transformers.image_utils import (
+    OPENAI_CLIP_MEAN,
+    OPENAI_CLIP_STD,
+    ChannelDimension,
+    PILImageResampling,
+    get_image_size,
+    infer_channel_dimension_format,
+    is_scaled_image,
+    is_valid_image,
+    make_list_of_images,
+    to_numpy_array,
+    valid_images,
+    validate_preprocess_arguments,
+)
+from transformers.utils import TensorType, is_vision_available, logging
+
+
+logger = logging.get_logger(__name__)
+
+
+if is_vision_available():
+    from PIL import Image
+
+ImageInput = Union[
+    "PIL.Image.Image",
+    np.ndarray,
+    "torch.Tensor",
+    List["PIL.Image.Image"],
+    List[np.ndarray],
+    List["torch.Tensor"],
+]  # noqa
+
+
+VideoInput = Union[
+    List["PIL.Image.Image"],
+    "np.ndarray",
+    "torch.Tensor",
+    List["np.ndarray"],
+    List["torch.Tensor"],
+    List[List["PIL.Image.Image"]],
+    List[List["np.ndarrray"]],
+    List[List["torch.Tensor"]],
+]  # noqa
+
+
+def make_batched_images(images) -> List[List[ImageInput]]:
+    """
+    Accepts images in list or nested list format, and makes a list of images for preprocessing.
+
+    Args:
+        images (`Union[List[List[ImageInput]], List[ImageInput], ImageInput]`):
+            The input image.
+
+    Returns:
+        list: A list of images.
+    """
+    if (
+        isinstance(images, (list, tuple))
+        and isinstance(images[0], (list, tuple))
+        and is_valid_image(images[0][0])
+    ):
+        return [img for img_list in images for img in img_list]
+
+    elif isinstance(images, (list, tuple)) and is_valid_image(images[0]):
+        return images
+
+    elif is_valid_image(images):
+        return [images]
+
+    raise ValueError(f"Could not make batched images from {images}")
+
+
+def adjust_size(size, patch_size):
+    num_patches = size // patch_size
+    if num_patches % 2 != 0:  # 如果是奇数，减1
+        num_patches -= 1
+    return num_patches * patch_size
+
+
+def make_batched_videos(videos) -> List[VideoInput]:
+    if (
+        isinstance(videos, (list, tuple))
+        and isinstance(videos[0], (list, tuple))
+        and is_valid_image(videos[0][0])
+    ):
+        return videos
+
+    elif isinstance(videos, (list, tuple)) and is_valid_image(videos[0]):
+        if isinstance(videos[0], Image.Image):
+            return [videos]
+        elif len(videos[0].shape) == 4:
+            return [list(video) for video in videos]
+
+    elif is_valid_image(videos) and len(videos.shape) == 4:
+        return [list(videos)]
+
+    raise ValueError(f"Could not make batched video from {videos}")
+
+
+def smart_resize(
+    height: int,
+    width: int,
+    factor: int = 28,
+    min_pixels: int = 28 * 28 * 130,
+    max_pixels: int = 28 * 28 * 1280,
+):
+    """Rescales the image so that the following conditions are met:
+
+    1. Both dimensions (height and width) are divisible by 'factor'.
+
+    2. The total number of pixels is within the range ['min_pixels', 'max_pixels'].
+
+    3. The aspect ratio of the image is maintained as closely as possible.
+
+    """
+    # if height < factor or width < factor:
+    #    raise ValueError(f"height:{height} or width:{width} must be larger than factor:{factor}")
+    # if int(height < factor//4) + int(width < factor//4):
+    #     raise ValueError(f"height:{height} or width:{width} must be larger than factor:{factor//4}")
+
+    if height < factor:
+        print(f"smart_resize: height={height} < factor={factor}, reset height=factor")
+        width = round((width * factor) / height)
+        height = factor
+
+    if width < factor:
+        print(f"smart_resize: width={width} < factor={factor}, reset width=factor")
+        height = round((height * factor) / width)
+        width = factor
+
+    if max(height, width) / min(height, width) > 200:
+        raise ValueError(
+            f"absolute aspect ratio must be smaller than 200, got {max(height, width) / min(height, width)}"
+        )
+    h_bar = round(height / factor) * factor
+    w_bar = round(width / factor) * factor
+    if h_bar * w_bar > max_pixels:
+        beta = math.sqrt((height * width) / max_pixels)
+        h_bar = math.floor(height / beta / factor) * factor
+        w_bar = math.floor(width / beta / factor) * factor
+    elif h_bar * w_bar < min_pixels:
+        beta = math.sqrt(min_pixels / (height * width))
+        h_bar = math.ceil(height * beta / factor) * factor
+        w_bar = math.ceil(width * beta / factor) * factor
+    return h_bar, w_bar
+
+
+class PaddleOCRVLImageProcessor(BaseImageProcessor):
+    r"""
+    Constructs a Siglip image processor that dynamically resizes images based on the original images.
+
+    Args:
+        do_resize (`bool`, *optional*, defaults to `True`):
+            Whether to resize the image's (height, width) dimensions.
+        resample (`PILImageResampling`, *optional*, defaults to `Resampling.BICUBIC`):
+            Resampling filter to use when resizing the image.
+        do_rescale (`bool`, *optional*, defaults to `True`):
+            Whether to rescale the image by the specified scale `rescale_factor`.
+        rescale_factor (`int` or `float`, *optional*, defaults to `1/255`):
+            Scale factor to use if rescaling the image.
+        do_normalize (`bool`, *optional*, defaults to `True`):
+            Whether to normalize the image.
+        image_mean (`float` or `List[float]`, *optional*, defaults to `[0.48145466, 0.4578275, 0.40821073]`):
+            Mean to use if normalizing the image. This is a float or list of floats for each channel in the image.
+        image_std (`float` or `List[float]`, *optional*, defaults to `[0.26862954, 0.26130258, 0.27577711]`):
+            Standard deviation to use if normalizing the image. This is a float or list of floats for each channel in the image.
+        do_convert_rgb (`bool`, *optional*, defaults to `True`):
+            Whether to convert the image to RGB.
+        min_pixels (`int`, *optional*, defaults to `28 * 28 * 130`):
+            The min pixels of the image to resize the image.
+        max_pixels (`int`, *optional*, defaults to `28 * 28 * 1670`):
+            The max pixels of the image to resize the image.
+        patch_size (`int`, *optional*, defaults to 14):
+            The spacial patch size of the vision encoder.
+        temporal_patch_size (`int`, *optional*, defaults to 2):
+            The temporal patch size of the vision encoder.
+        merge_size (`int`, *optional*, defaults to 2):
+            The merge size of the vision encoder to llm encoder.
+    """
+
+    model_input_names = [
+        "pixel_values",
+        "image_grid_thw",
+        "pixel_values_videos",
+        "video_grid_thw",
+    ]
+
+    def __init__(
+        self,
+        do_resize: bool = True,
+        resample: PILImageResampling = PILImageResampling.BICUBIC,
+        do_rescale: bool = True,
+        rescale_factor: Union[int, float] = 1 / 255,
+        do_normalize: bool = True,
+        image_mean: Optional[Union[float, List[float]]] = None,
+        image_std: Optional[Union[float, List[float]]] = None,
+        do_convert_rgb: bool = True,
+        min_pixels: int = 28 * 28 * 130,
+        max_pixels: int = 28 * 28 * 1280,
+        patch_size: int = 14,
+        temporal_patch_size: int = 1,
+        merge_size: int = 2,
+        **kwargs,
+    ) -> None:
+        super().__init__(**kwargs)
+        self.do_resize = do_resize
+        self.resample = resample
+        self.do_rescale = do_rescale
+        self.rescale_factor = rescale_factor
+        self.do_normalize = do_normalize
+        self.image_mean = image_mean if image_mean is not None else OPENAI_CLIP_MEAN
+        self.image_std = image_std if image_std is not None else OPENAI_CLIP_STD
+        self.min_pixels = min_pixels
+        self.max_pixels = max_pixels
+        self.patch_size = patch_size
+        self.temporal_patch_size = temporal_patch_size
+        self.merge_size = merge_size
+        self.size = {"min_pixels": min_pixels, "max_pixels": max_pixels}  # not used
+        self.do_convert_rgb = do_convert_rgb
+
+    def mvit_rescale(self, image: Image.Image, merge_size: int = 2) -> Image.Image:
+        try:
+            w, h = image.size
+        except:
+            raise ValueError(str((type(image), image)))
+        patch_size = self.patch_size
+
+        if (w // patch_size) * (h // patch_size) > self.in_token_limit:
+            scale = math.sqrt(
+                self.in_token_limit / ((w // patch_size) * (h // patch_size))
+            )
+            new_w, new_h = int(w * scale), int(h * scale)
+
+            image = image.resize((new_w, new_h), Image.Resampling.BICUBIC)
+        if self.pad_input:
+            new_w, new_h = image.size
+            pad_size_h = merge_size * patch_size
+            pad_size_w = merge_size * patch_size
+
+            pad_h = (pad_size_h - new_h % pad_size_h) % pad_size_h
+            pad_w = (pad_size_w - new_w % pad_size_w) % pad_size_w
+
+            image = TF.pad(image, (0, 0, pad_w, pad_h))
+        else:
+            new_w, new_h = image.size
+            new_w = new_w - new_w % patch_size
+            new_h = new_h - new_h % patch_size
+
+            new_w = adjust_size(new_w, patch_size)
+            new_h = adjust_size(new_h, patch_size)
+
+            image = TF.center_crop(image, (new_h, new_w))
+
+        w, h = image.size
+        if w // patch_size >= 512 or h // patch_size >= 512:
+            new_h = min(patch_size * 510, h)
+            new_w = min(patch_size * 510, w)
+            image = TF.center_crop(image, (new_h, new_w))
+            # raise ValueError("Exceed pos emb")
+        return image
+
+    def _preprocess(
+        self,
+        images: Union[ImageInput, VideoInput],
+        do_resize: bool = None,
+        resample: PILImageResampling = None,
+        do_rescale: bool = None,
+        rescale_factor: float = None,
+        do_normalize: bool = None,
+        image_mean: Optional[Union[float, List[float]]] = None,
+        image_std: Optional[Union[float, List[float]]] = None,
+        do_convert_rgb: bool = None,
+        data_format: Optional[ChannelDimension] = ChannelDimension.FIRST,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
+    ):
+        """
+        Preprocess an image or batch of images. Copy of the `preprocess` method from `CLIPImageProcessor`.
+
+        Args:
+            images (`ImageInput`):
+                Image or batch of images to preprocess. Expects pixel values ranging from 0 to 255. If pixel values range from 0 to 1, set `do_rescale=False`.
+            vision_info (`List[Dict]`, *optional*):
+                Optional list of dictionaries containing additional information about vision inputs.
+            do_resize (`bool`, *optional*, defaults to `self.do_resize`):
+                Whether to resize the image.
+            resample (`PILImageResampling`, *optional*, defaults to `self.resample`):
+                Resampling filter to use if resizing the image. This can be one of the `PILImageResampling` enums.
+            do_rescale (`bool`, *optional*, defaults to `self.do_rescale`):
+                Whether to rescale the image.
+            rescale_factor (`float`, *optional*, defaults to `self.rescale_factor`):
+                Scale factor to use if rescaling the image.
+            do_normalize (`bool`, *optional*, defaults to `self.do_normalize`):
+                Whether to normalize the image.
+            image_mean (`float` or `List[float]`, *optional*, defaults to `self.image_mean`):
+                Mean to use if normalizing the image. Can be a float or a list of floats corresponding to the number of channels in the image.
+            image_std (`float` or `List[float]`, *optional*, defaults to `self.image_std`):
+                Standard deviation to use if normalizing the image. Can be a float or a list of floats corresponding to the number of channels in the image.
+            do_convert_rgb (`bool`, *optional*, defaults to `self.do_convert_rgb`):
+                Whether to convert the image to RGB.
+            data_format (`ChannelDimension`, *optional*, defaults to `ChannelDimension.FIRST`):
+                The channel dimension format for the output image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - Unset: Use the channel dimension format of the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.   - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
+        """
+        images = make_list_of_images(images)
+
+        if input_data_format is None:
+            # We assume that all images have the same channel dimension format.
+            input_data_format = ChannelDimension.LAST if isinstance(images[0], Image.Image) else infer_channel_dimension_format(images[0])
+
+        if do_convert_rgb:
+            images = [convert_to_rgb(image) for image in images]
+
+        # All transformations expect numpy arrays.
+        images = [to_numpy_array(image) for image in images]
+
+        if is_scaled_image(images[0]) and do_rescale:
+            logger.warning_once(
+                "It looks like you are trying to rescale already rescaled images. If the input"
+                " images have pixel values between 0 and 1, set `do_rescale=False` to avoid rescaling them again."
+            )
+
+        height, width = get_image_size(images[0], channel_dim=input_data_format)
+        resized_height, resized_width = height, width
+        processed_images = []
+
+        for image in images:
+            if do_resize:
+                resized_height, resized_width = smart_resize(
+                    height,
+                    width,
+                    factor=self.patch_size * self.merge_size,
+                    min_pixels=self.min_pixels,
+                    max_pixels=self.max_pixels,
+                )
+                image = resize(
+                    image,
+                    size=(resized_height, resized_width),
+                    resample=resample,
+                    input_data_format=input_data_format,
+                )
+
+            if do_rescale:
+                image = self.rescale(
+                    image, scale=rescale_factor, input_data_format=input_data_format
+                )
+
+            if do_normalize:
+                image = self.normalize(
+                    image=image,
+                    mean=image_mean,
+                    std=image_std,
+                    input_data_format=input_data_format,
+                )
+            image = to_channel_dimension_format(
+                image, data_format, input_channel_dim=input_data_format
+            )
+            processed_images.append(image)
+
+        patches = np.array(processed_images)
+        if data_format == ChannelDimension.LAST:
+            patches = patches.transpose(0, 3, 1, 2)
+        if patches.shape[0] == 1:
+            patches = np.tile(patches, (self.temporal_patch_size, 1, 1, 1))
+        init_patches = patches
+        channel = patches.shape[1]
+        grid_t = patches.shape[0] // self.temporal_patch_size
+        grid_h, grid_w = (
+            resized_height // self.patch_size,
+            resized_width // self.patch_size,
+        )
+        patches = patches.reshape(
+            grid_t,
+            self.temporal_patch_size,
+            channel,
+            grid_h,
+            self.patch_size,
+            grid_w,
+            self.patch_size,
+        )
+        patches = patches.transpose(0, 3, 5, 2, 1, 4, 6)
+        assert self.temporal_patch_size == 1
+        flatten_patches = patches.reshape(
+            grid_t * grid_h * grid_w, channel, self.patch_size, self.patch_size
+        )
+        return flatten_patches, (grid_t, grid_h, grid_w)
+
+    def preprocess(
+        self,
+        images: ImageInput,
+        videos: VideoInput = None,
+        do_resize: bool = None,
+        size: Dict[str, int] = None,
+        resample: PILImageResampling = None,
+        do_rescale: bool = None,
+        rescale_factor: float = None,
+        do_normalize: bool = None,
+        image_mean: Optional[Union[float, List[float]]] = None,
+        image_std: Optional[Union[float, List[float]]] = None,
+        do_convert_rgb: bool = None,
+        return_tensors: Optional[Union[str, TensorType]] = None,
+        data_format: Optional[ChannelDimension] = ChannelDimension.FIRST,
+        input_data_format: Optional[Union[str, ChannelDimension]] = None,
+    ):
+        """
+        Args:
+            images (`ImageInput`):
+                Image to preprocess. Expects a single or batch of images with pixel values ranging from 0 to 255. If
+                passing in images with pixel values between 0 and 1, set `do_rescale=False`.
+            videos (`VideoInput`):
+                Video to preprocess. Expects a single or batch of videos with pixel values ranging from 0 to 255. If
+                passing in videos with pixel values between 0 and 1, set `do_rescale=False`.
+            do_resize (`bool`, *optional*, defaults to `self.do_resize`):
+                Whether to resize the image.
+            size (`Dict[str, int]`, *optional*, defaults to `self.size`):
+                Size of the image after resizing. Shortest edge of the image is resized to size["shortest_edge"], with
+                the longest edge resized to keep the input aspect ratio.
+            resample (`int`, *optional*, defaults to `self.resample`):
+                Resampling filter to use if resizing the image. This can be one of the enum `PILImageResampling`. Only
+                has an effect if `do_resize` is set to `True`.
+            do_rescale (`bool`, *optional*, defaults to `self.do_rescale`):
+                Whether to rescale the image.
+            rescale_factor (`float`, *optional*, defaults to `self.rescale_factor`):
+                Rescale factor to rescale the image by if `do_rescale` is set to `True`.
+            do_normalize (`bool`, *optional*, defaults to `self.do_normalize`):
+                Whether to normalize the image.
+            image_mean (`float` or `List[float]`, *optional*, defaults to `self.image_mean`):
+                Image mean to use for normalization. Only has an effect if `do_normalize` is set to `True`.
+            image_std (`float` or `List[float]`, *optional*, defaults to `self.image_std`):
+                Image standard deviation to use for normalization. Only has an effect if `do_normalize` is set to
+                `True`.
+            do_convert_rgb (`bool`, *optional*, defaults to `self.do_convert_rgb`):
+                Whether to convert the image to RGB.
+            return_tensors (`str` or `TensorType`, *optional*):
+                The type of tensors to return. Can be one of:
+                - Unset: Return a list of `np.ndarray`.
+                - `TensorType.TENSORFLOW` or `'tf'`: Return a batch of type `tf.Tensor`.
+                - `TensorType.PYTORCH` or `'pt'`: Return a batch of type `torch.Tensor`.
+                - `TensorType.NUMPY` or `'np'`: Return a batch of type `np.ndarray`.
+                - `TensorType.JAX` or `'jax'`: Return a batch of type `jax.numpy.ndarray`.
+            data_format (`ChannelDimension` or `str`, *optional*, defaults to `ChannelDimension.FIRST`):
+                The channel dimension format for the output image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - Unset: Use the channel dimension format of the input image.
+            input_data_format (`ChannelDimension` or `str`, *optional*):
+                The channel dimension format for the input image. If unset, the channel dimension format is inferred
+                from the input image. Can be one of:
+                - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
+                - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format.
+                - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.
+
+        """
+        do_resize = do_resize if do_resize is not None else self.do_resize
+        size = size if size is not None else self.size
+        resample = resample if resample is not None else self.resample
+        do_rescale = do_rescale if do_rescale is not None else self.do_rescale
+        rescale_factor = (
+            rescale_factor if rescale_factor is not None else self.rescale_factor
+        )
+        do_normalize = do_normalize if do_normalize is not None else self.do_normalize
+        image_mean = image_mean if image_mean is not None else self.image_mean
+        image_std = image_std if image_std is not None else self.image_std
+        do_convert_rgb = (
+            do_convert_rgb if do_convert_rgb is not None else self.do_convert_rgb
+        )
+
+        if images is not None:
+            images = make_batched_images(images)
+        if videos is not None:
+            videos = make_batched_videos(videos)
+
+        if images is not None and not valid_images(images):
+            raise ValueError(
+                "Invalid image type. Must be of type PIL.Image.Image, numpy.ndarray, "
+                "torch.Tensor, tf.Tensor or jax.ndarray."
+            )
+
+        validate_preprocess_arguments(
+            rescale_factor=rescale_factor,
+            do_normalize=do_normalize,
+            image_mean=image_mean,
+            image_std=image_std,
+            do_resize=do_resize,
+            size=size,
+            resample=resample,
+        )
+
+        if images is not None:
+            pixel_values, vision_grid_thws = [], []
+            for image in images:
+                patches, image_grid_thw = self._preprocess(
+                    image,
+                    do_resize=do_resize,
+                    resample=resample,
+                    do_rescale=do_rescale,
+                    rescale_factor=rescale_factor,
+                    do_normalize=do_normalize,
+                    image_mean=image_mean,
+                    image_std=image_std,
+                    data_format=data_format,
+                    do_convert_rgb=do_convert_rgb,
+                    input_data_format=input_data_format,
+                )
+                pixel_values.extend(patches)
+                vision_grid_thws.append(image_grid_thw)
+            pixel_values = np.array(pixel_values)
+            vision_grid_thws = np.array(vision_grid_thws)
+            data = {"pixel_values": pixel_values, "image_grid_thw": vision_grid_thws}
+
+        if videos is not None:
+            pixel_values, vision_grid_thws = [], []
+            for images in videos:
+                patches, video_grid_thw = self._preprocess(
+                    images,
+                    do_resize=do_resize,
+                    resample=resample,
+                    do_rescale=do_rescale,
+                    rescale_factor=rescale_factor,
+                    do_normalize=do_normalize,
+                    image_mean=image_mean,
+                    image_std=image_std,
+                    data_format=data_format,
+                    do_convert_rgb=do_convert_rgb,
+                    input_data_format=input_data_format,
+                )
+                pixel_values.extend(patches)
+                vision_grid_thws.append(video_grid_thw)
+            pixel_values = np.array(pixel_values)
+            vision_grid_thws = np.array(vision_grid_thws)
+            data = {
+                "pixel_values_videos": pixel_values,
+                "video_grid_thw": vision_grid_thws,
+            }
+
+        return BatchFeature(data=data, tensor_type=return_tensors)

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d1b2613b26009260e1f98cf691f14da9f9051ad3880a172cc428d1f7f05934e7
+size 1811280336

processing_paddleocr_vl.py

@@ -0,0 +1,293 @@
+# Copyright (c) 2025 PaddlePaddle Authors. All Rights Reserved.
+#
+# 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.
+
+from typing import List, Union
+import numpy as np
+import torch
+from transformers.feature_extraction_utils import BatchFeature
+from transformers.processing_utils import (
+    ProcessingKwargs,
+    ProcessorMixin,
+    Unpack,
+    VideosKwargs,
+)
+from transformers.tokenization_utils_base import PreTokenizedInput, TextInput
+
+
+ImageInput = Union[
+    "PIL.Image.Image",
+    np.ndarray,
+    "torch.Tensor",
+    List["PIL.Image.Image"],
+    List[np.ndarray],
+    List["torch.Tensor"],
+]  # noqa
+
+
+VideoInput = Union[
+    List["PIL.Image.Image"],
+    "np.ndarray",
+    "torch.Tensor",
+    List["np.ndarray"],
+    List["torch.Tensor"],
+    List[List["PIL.Image.Image"]],
+    List[List["np.ndarrray"]],
+    List[List["torch.Tensor"]],
+]  # noqa
+
+
+class PaddleOCRVLVideosProcessorKwargs(VideosKwargs, total=False):
+    fps: Union[List[float], float]
+
+
+class PaddleOCRVLProcessorKwargs(ProcessingKwargs, total=False):
+    videos_kwargs: PaddleOCRVLVideosProcessorKwargs
+    _defaults = {
+        "text_kwargs": {
+            "padding": False,
+        },
+        "videos_kwargs": {"fps": 2.0},
+    }
+
+
+class PaddleOCRVLProcessor(ProcessorMixin):
+    r"""
+    [`PaddleOCRVLProcessor`] offers all the functionalities of [`SiglipImageProcessor`] and [`Qwen2TokenizerFast`]. See the
+    [`~PaddleOCRVLProcessor.__call__`] and [`~PaddleOCRVLProcessor.decode`] for more information.
+    Args:
+        image_processor ([`SiglipImageProcessor`], *optional*):
+            The image processor is a required input.
+        tokenizer ([`Qwen2TokenizerFast`], *optional*):
+            The tokenizer is a required input.
+        chat_template (`str`, *optional*): A Jinja template which will be used to convert lists of messages
+            in a chat into a tokenizable string.
+    """
+
+    attributes = ["image_processor", "tokenizer"]
+    valid_kwargs = [
+        "chat_template",
+        "image_std",
+        "min_pixels",
+        "image_mean",
+        "merge_size",
+        "image_processor_type",
+        "temporal_patch_size",
+        "patch_size",
+        "max_pixels",
+    ]
+
+    image_processor_class = "AutoImageProcessor"
+    tokenizer_class = "AutoTokenizer"
+
+    def __init__(
+        self, image_processor=None, tokenizer=None, chat_template=None, **kwargs
+    ):
+        self.image_token = (
+            "<|IMAGE_PLACEHOLDER|>"
+            if not hasattr(tokenizer, "image_token")
+            else tokenizer.image_token
+        )
+        self.video_token = (
+            "<|video_pad|>"
+            if not hasattr(tokenizer, "video_token")
+            else tokenizer.video_token
+        )
+        super().__init__(image_processor, tokenizer, chat_template=chat_template)
+
+    def __call__(
+        self,
+        images: ImageInput = None,
+        text: Union[
+            TextInput, PreTokenizedInput, List[TextInput], List[PreTokenizedInput]
+        ] = None,
+        videos: VideoInput = None,
+        **kwargs: Unpack[PaddleOCRVLProcessorKwargs],
+    ) -> BatchFeature:
+        """
+        Main method to prepare for the model one or several sequences(s) and image(s). This method forwards the `text`
+        and `kwargs` arguments to Qwen2TokenizerFast's [`~Qwen2TokenizerFast.__call__`] if `text` is not `None` to encode
+        the text. To prepare the vision inputs, this method forwards the `vision_infos` and `kwrags` arguments to
+        SiglipImageProcessor's [`~SiglipImageProcessor.__call__`] if `vision_infos` is not `None`.
+
+        Args:
+            images (`PIL.Image.Image`, `np.ndarray`, `torch.Tensor`, `List[PIL.Image.Image]`, `List[np.ndarray]`, `List[torch.Tensor]`):
+                The image or batch of images to be prepared. Each image can be a PIL image, NumPy array or PyTorch
+                tensor. Both channels-first and channels-last formats are supported.
+            text (`str`, `List[str]`, `List[List[str]]`):
+                The sequence or batch of sequences to be encoded. Each sequence can be a string or a list of strings
+                (pretokenized string). If the sequences are provided as list of strings (pretokenized), you must set
+                `is_split_into_words=True` (to lift the ambiguity with a batch of sequences).
+            videos (`np.ndarray`, `torch.Tensor`, `List[np.ndarray]`, `List[torch.Tensor]`):
+                The image or batch of videos to be prepared. Each video can be a 4D NumPy array or PyTorch
+                tensor, or a nested list of 3D frames. Both channels-first and channels-last formats are supported.
+            return_tensors (`str` or [`~utils.TensorType`], *optional*):
+                If set, will return tensors of a particular framework. Acceptable values are:
+                - `'tf'`: Return TensorFlow `tf.constant` objects.
+                - `'pt'`: Return PyTorch `torch.Tensor` objects.
+                - `'np'`: Return NumPy `np.ndarray` objects.
+                - `'jax'`: Return JAX `jnp.ndarray` objects.
+
+        Returns:
+            [`BatchFeature`]: A [`BatchFeature`] with the following fields:
+
+            - **input_ids** -- List of token ids to be fed to a model. Returned when `text` is not `None`.
+            - **attention_mask** -- List of indices specifying which tokens should be attended to by the model (when
+              `return_attention_mask=True` or if *"attention_mask"* is in `self.model_input_names` and if `text` is not
+              `None`).
+            - **pixel_values** -- Pixel values to be fed to a model. Returned when `images` is not `None`.
+            - **pixel_values_videos** -- Pixel values of videos to be fed to a model. Returned when `videos` is not `None`.
+            - **image_grid_thw** -- List of image 3D grid in LLM. Returned when `images` is not `None`.
+            - **video_grid_thw** -- List of video 3D grid in LLM. Returned when `videos` is not `None`.
+            - **second_per_grid_ts** -- List of video seconds per time grid. Returned when `videos` is not `None`.
+        """
+        output_kwargs = self._merge_kwargs(
+            PaddleOCRVLProcessorKwargs,
+            tokenizer_init_kwargs=self.tokenizer.init_kwargs,
+            **kwargs,
+        )
+
+        if images is not None:
+            image_inputs = self.image_processor(images=images, return_tensors="pt")
+            image_inputs["pixel_values"] = image_inputs["pixel_values"]
+            image_grid_thw = image_inputs["image_grid_thw"]
+
+        else:
+            image_inputs = {}
+            image_grid_thw = None
+
+        if videos is not None:
+            # TODO: add video processing
+            videos_inputs = self.image_processor(
+                images=None, videos=videos, **output_kwargs["images_kwargs"]
+            )
+            video_grid_thw = videos_inputs["video_grid_thw"]
+
+            fps = output_kwargs["videos_kwargs"].pop("fps", 2.0)
+            if isinstance(fps, (int, float)):
+                second_per_grid_ts = [
+                    self.image_processor.temporal_patch_size / fps
+                ] * len(video_grid_thw)
+            elif hasattr(fps, "__len__") and len(fps) == len(video_grid_thw):
+                second_per_grid_ts = [
+                    self.image_processor.temporal_patch_size / tmp for tmp in fps
+                ]
+            else:
+                raise ValueError(
+                    f"The length of fps ({len(fps) if hasattr(fps, '__len__') else fps}) must be equal to the length of video_grid_thw ({len(video_grid_thw)}) or fps should be a single number."
+                )
+            videos_inputs.update(
+                {"second_per_grid_ts": torch.tensor(second_per_grid_ts)}
+            )
+
+        else:
+            videos_inputs = {}
+            video_grid_thw = None
+
+        if not isinstance(text, list):
+            text = [text]
+
+        if image_grid_thw is not None:
+            index = 0
+            for i in range(len(text)):
+                while self.image_token in text[i]:
+                    text[i] = text[i].replace(
+                        self.image_token,
+                        "<|placeholder|>"
+                        * (
+                            image_grid_thw[index].prod()
+                            // self.image_processor.merge_size
+                            // self.image_processor.merge_size
+                        ),
+                        1,
+                    )
+                    index += 1
+                text[i] = text[i].replace("<|placeholder|>", self.image_token)
+
+        if video_grid_thw is not None:
+            index = 0
+            for i in range(len(text)):
+                while self.video_token in text[i]:
+                    text[i] = text[i].replace(
+                        self.video_token,
+                        "<|placeholder|>"
+                        * (
+                            video_grid_thw[index].prod()
+                            // self.image_processor.merge_size
+                            // self.image_processor.merge_size
+                        ),
+                        1,
+                    )
+                    index += 1
+                text[i] = text[i].replace("<|placeholder|>", self.video_token)
+
+        text_inputs = self.tokenizer(text, **output_kwargs["text_kwargs"])
+
+        return BatchFeature(data={**text_inputs, **image_inputs, **videos_inputs})
+
+    def batch_decode(self, *args, **kwargs):
+        """
+        This method forwards all its arguments to Qwen2TokenizerFast's [`~PreTrainedTokenizer.batch_decode`]. Please
+        refer to the docstring of this method for more information.
+        """
+        return self.tokenizer.batch_decode(*args, **kwargs)
+
+    def decode(self, *args, **kwargs):
+        """
+        This method forwards all its arguments to Qwen2TokenizerFast's [`~PreTrainedTokenizer.decode`]. Please refer to
+        the docstring of this method for more information.
+        """
+        return self.tokenizer.decode(*args, **kwargs)
+
+    def post_process_image_text_to_text(
+        self,
+        generated_outputs,
+        skip_special_tokens=True,
+        clean_up_tokenization_spaces=False,
+        **kwargs,
+    ):
+        """
+        Post-process the output of the model to decode the text.
+
+        Args:
+            generated_outputs (`torch.Tensor` or `np.ndarray`):
+                The output of the model `generate` function. The output is expected to be a tensor of shape `(batch_size, sequence_length)`
+                or `(sequence_length,)`.
+            skip_special_tokens (`bool`, *optional*, defaults to `True`):
+                Whether or not to remove special tokens in the output. Argument passed to the tokenizer's `batch_decode` method.
+            Clean_up_tokenization_spaces (`bool`, *optional*, defaults to `False`):
+                Whether or not to clean up the tokenization spaces. Argument passed to the tokenizer's `batch_decode` method.
+            **kwargs:
+                Additional arguments to be passed to the tokenizer's `batch_decode method`.
+
+        Returns:
+            `List[str]`: The decoded text.
+        """
+        return self.tokenizer.batch_decode(
+            generated_outputs,
+            skip_special_tokens=skip_special_tokens,
+            clean_up_tokenization_spaces=clean_up_tokenization_spaces,
+            **kwargs,
+        )
+
+    @property
+    def model_input_names(self):
+        tokenizer_input_names = self.tokenizer.model_input_names
+        image_processor_input_names = self.image_processor.model_input_names
+        names_from_processor = list(
+            dict.fromkeys(tokenizer_input_names + image_processor_input_names)
+        )
+        return names_from_processor + ["second_per_grid_ts"]
+
+
+__all__ = ["PaddleOCRVLProcessor", "PaddleOCRVLProcessor"]

processor_config.json

@@ -0,0 +1,33 @@
+{
+  "image_processor": {
+    "auto_map": {
+      "AutoImageProcessor": "image_processing_paddleocr_vl.PaddleOCRVLImageProcessor",
+      "AutoProcessor": "processing_paddleocr_vl.PaddleOCRVLProcessor"
+    },
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "do_resize": true,
+    "image_mean": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "image_processor_type": "PaddleOCRVLImageProcessor",
+    "image_std": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "merge_size": 2,
+    "patch_size": 14,
+    "resample": 3,
+    "rescale_factor": 0.00392156862745098,
+    "size": {
+      "longest_edge": 1003520,
+      "shortest_edge": 112896
+    },
+    "temporal_patch_size": 1
+  },
+  "processor_class": "PaddleOCRVLProcessor"
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c8a215a59183d0d0781adc33bacd3ce6162716f7fd568fb30234a74d69803a7d
+size 11189060

tokenizer_config.json

@@ -0,0 +1,28 @@
+{
+  "add_prefix_space": false,
+  "auto_map": {
+    "AutoProcessor": "processing_paddleocr_vl.PaddleOCRVLProcessor"
+  },
+  "backend": "tokenizers",
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "cls_token": "<|begin_of_sentence|>",
+  "eos_token": "</s>",
+  "image_token": "<|IMAGE_PLACEHOLDER|>",
+  "is_local": false,
+  "legacy": true,
+  "local_files_only": false,
+  "mask_token": "<mask:1>",
+  "model_max_length": 131072,
+  "model_specific_special_tokens": {
+    "image_token": "<|IMAGE_PLACEHOLDER|>"
+  },
+  "pad_token": "<unk>",
+  "processor_class": "PaddleOCRVLProcessor",
+  "sep_token": "<|end_of_sentence|>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "TokenizersBackend",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

usage.py

@@ -0,0 +1,53 @@
+"""Read one electrical one-line diagram image into JSON.
+
+    python usage.py drawing.png [--device cuda|cpu]
+
+Apache-2.0, Copyright 2026 Decosa.
+"""
+import argparse
+import json
+import re
+
+import torch
+from PIL import Image
+from transformers import AutoModelForImageTextToText, AutoProcessor
+
+REPO = "decosaai/decosa-oneline-reader-paddleocr-vl"
+PROMPT = "One-line Diagram Recognition:"
+MAX_PIXELS = 1003520  # the processor's cap
+
+
+def fit(im: Image.Image) -> Image.Image:
+    """Scale to just under the processor's pixel cap; small or poor scans are enlarged (as in training)."""
+    w, h = im.size
+    s = (MAX_PIXELS * 0.98 / (w * h)) ** 0.5
+    if s < 1 or w < 1100:
+        im = im.resize((max(28, int(w * s)), max(28, int(h * s))), Image.BICUBIC)
+    return im
+
+
+def main() -> None:
+    ap = argparse.ArgumentParser()
+    ap.add_argument("image")
+    ap.add_argument("--model", default=REPO)
+    ap.add_argument("--device", default="cuda" if torch.cuda.is_available() else "cpu")
+    a = ap.parse_args()
+    proc = AutoProcessor.from_pretrained(a.model)
+    dtype = torch.bfloat16 if a.device == "cuda" else torch.float32
+    model = AutoModelForImageTextToText.from_pretrained(a.model, dtype=dtype).to(a.device).eval()
+    im = fit(Image.open(a.image).convert("L")).convert("RGB")
+    msgs = [{"role": "user", "content": [{"type": "image", "image": im}, {"type": "text", "text": PROMPT}]}]
+    x = proc.apply_chat_template(msgs, add_generation_prompt=True, tokenize=True, return_dict=True,
+                                 return_tensors="pt").to(a.device)
+    with torch.inference_mode():
+        out = model.generate(**x, max_new_tokens=900, do_sample=False, use_cache=True)
+    text = proc.decode(out[0][x["input_ids"].shape[-1]:], skip_special_tokens=True)
+    text = re.sub(r"<\|[a-z_]+\|>|</s>|<s>|<pad>", "", text).strip()
+    try:
+        print(json.dumps(json.loads(text), indent=1))
+    except json.JSONDecodeError:
+        print(text)
+
+
+if __name__ == "__main__":
+    main()