Update README.md

README.md

@@ -3,7 +3,6 @@ license: apache-2.0
 pipeline_tag: visual-document-retrieval
 library_name: transformers
 ---
-
 <div align="center" xmlns="http://www.w3.org/1999/html">
 <h1 align="center">
 MonkeyOCR: Document Parsing with a Structure-Recognition-Relation Triplet Paradigm
@@ -21,7 +20,8 @@ MonkeyOCR: Document Parsing with a Structure-Recognition-Relation Triplet Paradi
 > Zhang Li, Yuliang Liu, Qiang Liu, Zhiyin Ma, Ziyang Zhang, Shuo Zhang, Zidun Guo, Jiarui Zhang, Xinyu Wang, Xiang Bai <br>
 [![arXiv](https://img.shields.io/badge/Arxiv-b31b1b.svg?logo=arXiv)](https://arxiv.org/abs/2506.05218) 
 [![Source_code](https://img.shields.io/badge/Code-Available-white)](README.md)
-[![Model Weight](https://img.shields.io/badge/Model_Weight-gray)](https://huggingface.co/echo840/MonkeyOCR)
+[![Model Weight](https://img.shields.io/badge/HuggingFace-gray)](https://huggingface.co/echo840/MonkeyOCR)
+[![Model Weight](https://img.shields.io/badge/ModelScope-green)](https://modelscope.cn/models/l1731396519/MonkeyOCR)
 [![Demo](https://img.shields.io/badge/Demo-blue)](http://vlrlabmonkey.xyz:7685/)
 
 
@@ -33,9 +33,11 @@ MonkeyOCR adopts a Structure-Recognition-Relation (SRR) triplet paradigm, which
 2. Compared to end-to-end models, our 3B-parameter model achieves the best average performance on English documents, outperforming models such as Gemini 2.5 Pro and Qwen2.5 VL-72B.
 3. For multi-page document parsing, our method reaches a processing speed of 0.84 pages per second, surpassing MinerU (0.65) and Qwen2.5 VL-7B (0.12).
 
-
 <img src="https://v1.ax1x.com/2025/06/05/7jQ3cm.png" alt="7jQ3cm.png" border="0" />
 
+MonkeyOCR currently does not support photographed documents, but we will continue to improve it in future updates. Stay tuned!
+Currently, our model is deployed on a single GPU, so if too many users upload files at the same time, issues like “This application is currently busy” may occur. We're actively working on supporting Ollama and other deployment solutions to ensure a smoother experience for more users. Additionally, please note that the processing time shown on the demo page does not reflect computation time alone—it also includes result uploading and other overhead. During periods of high traffic, this time may be longer. The inference speeds of MonkeyOCR, MinerU, and Qwen2.5 VL-7B were measured on an H800 GPU.
+
 ## News 
 * ```2025.06.05 ``` 🚀 We release MonkeyOCR, which supports the parsing of various types of Chinese and English documents.
 
@@ -52,23 +54,45 @@ cd MonkeyOCR
 
 # Install pytorch, see https://pytorch.org/get-started/previous-versions/ for your cuda version
 pip install torch==2.5.1 torchvision==0.20.1 torchaudio==2.5.1 --index-url https://download.pytorch.org/whl/cu124 
-pip install .
+pip install -e .
 ```
 ### 2. Download Model Weights
+Download our model from Huggingface.
 ```python
 pip install huggingface_hub
 
-python download_model.py
+python tools/download_model.py
+```
+You can also download our model from ModelScope.
 
+```python
+pip install modelscope
+
+python tools/download_model.py -t modelscope
 ```
 ### 3. Inference
 ```bash
 # Make sure in MonkeyOCR directory
 python parse.py path/to/your.pdf
-# Specify MonkeyChat path and model configs path
-python parse.py path/to/your.pdf -m model_weight/Recognition -c config.yaml
+# or with image as input
+pyhton parse.py path/to/your/image
+# Specify output path and model configs path
+python parse.py path/to/your.pdf -o ./output -c config.yaml
 ```
-### 4. Gradio demo
+
+#### Output Results
+MonkeyOCR generates three types of output files:
+
+1. **Processed Markdown File** (`your.md`): The final parsed document content in markdown format, containing text, formulas, tables, and other structured elements.
+2. **Layout Results** (`your_layout.pdf`): The layout results drawed on origin PDF.
+2. **Intermediate Block Results** (`your_middle.json`): A JSON file containing detailed information about all detected blocks, including:
+   - Block coordinates and positions
+   - Block content and type information
+   - Relationship information between blocks
+
+These files provide both the final formatted output and detailed intermediate results for further analysis or processing.
+
+### 4. Gradio Demo
 ```bash
 # Prepare your env for gradio
 pip install gradio==5.23.3
@@ -78,8 +102,49 @@ pip install pdf2image==1.17.0
 # Start demo
 python demo/demo_gradio.py
 ```
+### Fix **shared memory error** on **RTX 3090 / 4090 / ...** GPUs (Optional)
+
+Our 3B model runs efficiently on NVIDIA RTX 3090. However, when using **LMDeploy** as the inference backend, you may encounter compatibility issues on **RTX 3090 / 4090** GPUs — particularly the following error:
+
+```
+triton.runtime.errors.OutOfResources: out of resource: shared memory
+```
+
+To work around this issue, you can apply the patch below:
 
-Using the [LMDeploy](https://github.com/InternLM/lmdeploy), our model can run efficiently on an NVIDIA 3090 GPU.
+```bash
+python tools/lmdeploy_patcher.py patch
+```
+
+> ⚠️ **Note:** This command will modify LMDeploy's source code in your environment.
+> To revert the changes, simply run:
+
+```bash
+python tools/lmdeploy_patcher.py restore
+```
+
+**Special thanks to [@pineking](https://github.com/pineking) for the solution!**
+
+### Switch inference backend (Optional)
+
+You can switch inference backend to `transformers` following the steps below:
+
+1. Install required dependency (if not already installed):
+   ```bash
+   # install flash attention 2, you can download the corresponding version from https://github.com/Dao-AILab/flash-attention/releases/
+   pip install flash-attn==2.7.4.post1 --no-build-isolation
+   ```
+2. Open the `model_configs.yaml` file
+3. Set `chat_config.backend` to `transformers`
+4. Adjust the `batch_size` according to your GPU's memory capacity to ensure stable performance
+
+Example configuration:
+
+```yaml
+chat_config:
+  backend: transformers
+  batch_size: 10  # Adjust based on your available GPU memory
+```
 
 
 ## Benchmark Results
@@ -582,15 +647,29 @@ Here are the evaluation results of our model on OmniDocBench. MonkeyOCR-3B uses
 
 ## Visualization Demo
 
-Demo Link:  http://vlrlabmonkey.xyz:7685
+Get a Quick Hands-On Experience with Our Demo:  http://vlrlabmonkey.xyz:7685
+
 > Our demo is simple and easy to use:
 >
 > 1. Upload a PDF or image.
 > 2. Click “Parse (解析)” to let the model perform structure detection, content recognition, and relationship prediction on the input document. The final output will be a markdown-formatted version of the document.
-> 3. Select a prompt and click “Chat (对话)” to let the model perform content recognition on the image based on the selected prompt.
+> 3. Select a prompt and click “Test by prompt” to let the model perform content recognition on the image based on the selected prompt.
+
+
+
+
+### Example for formula document
+<img src="https://v1.ax1x.com/2025/06/10/7jVLgB.jpg" alt="7jVLgB.jpg" border="0" />
 
+### Example for table document
+<img src="https://v1.ax1x.com/2025/06/11/7jcOaa.png" alt="7jcOaa.png" border="0" />
 
+### Example for newspaper
+<img src="https://v1.ax1x.com/2025/06/11/7jcP5V.png" alt="7jcP5V.png" border="0" />
 
+### Example for financial report
+<img src="https://v1.ax1x.com/2025/06/11/7jc10I.png" alt="7jc10I.png" border="0" />
+<img src="https://v1.ax1x.com/2025/06/11/7jcRCL.png" alt="7jcRCL.png" border="0" />
 
 ## Citing MonkeyOCR
 
@@ -615,4 +694,4 @@ We would like to thank [MinerU](https://github.com/opendatalab/MinerU), [DocLayo
 
 
 ## Copyright
-MonkeyDoc dataset was collected from public datasets, crawled from the internet, and obtained through our own photography. The current technical report only presents the results of the 3B model. If you are interested in larger one, please contact Prof. Yuliang Liu at [email protected].
+Please don’t hesitate to share your valuable feedback — it’s a key motivation that drives us to continuously improve our framework. The current technical report only presents the results of the 3B model. Our model is intended for non-commercial use. If you are interested in larger one, please contact us at [email protected] or [email protected].