Image-Text-to-Text
Transformers
Safetensors
multilingual
GOT
feature-extraction
got
vision-language
ocr2.0
custom_code
Instructions to use Mageia/GOT-OCR2_0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Mageia/GOT-OCR2_0 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="Mageia/GOT-OCR2_0", trust_remote_code=True)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Mageia/GOT-OCR2_0", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use Mageia/GOT-OCR2_0 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Mageia/GOT-OCR2_0" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Mageia/GOT-OCR2_0", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/Mageia/GOT-OCR2_0
- SGLang
How to use Mageia/GOT-OCR2_0 with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "Mageia/GOT-OCR2_0" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Mageia/GOT-OCR2_0", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "Mageia/GOT-OCR2_0" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Mageia/GOT-OCR2_0", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use Mageia/GOT-OCR2_0 with Docker Model Runner:
docker model run hf.co/Mageia/GOT-OCR2_0
| punctuation_dict = {",": ",", "。": "."} | |
| translation_table = str.maketrans(punctuation_dict) | |
| def svg_to_html(svg_content, output_filename): | |
| html_content = f""" | |
| <!DOCTYPE html> | |
| <html lang="en"> | |
| <head> | |
| <meta charset="UTF-8"> | |
| <meta name="viewport" content="width=device-width, initial-scale=1.0"> | |
| <title>SVG Embedded in HTML</title> | |
| </head> | |
| <body> | |
| <svg width="2100" height="15000" xmlns="http://www.w3.org/2000/svg"> | |
| {svg_content} | |
| </svg> | |
| </body> | |
| </html> | |
| """ | |
| with open(output_filename, "w") as file: | |
| file.write(html_content) | |
| content_mmd_to_html = """<!DOCTYPE html> | |
| <html lang="en" data-lt-installed="true"><head> | |
| <meta charset="UTF-8"> | |
| <title>Title</title> | |
| <script> | |
| const text = | |
| </script> | |
| <style> | |
| #content { | |
| max-width: 800px; | |
| margin: auto; | |
| } | |
| </style> | |
| <script> | |
| let script = document.createElement('script'); | |
| script.src = "https://cdn.jsdelivr.net/npm/mathpix-markdown-it@1.3.6/es5/bundle.js"; | |
| document.head.append(script); | |
| script.onload = function() { | |
| const isLoaded = window.loadMathJax(); | |
| if (isLoaded) { | |
| console.log('Styles loaded!') | |
| } | |
| const el = window.document.getElementById('content-text'); | |
| if (el) { | |
| const options = { | |
| htmlTags: true | |
| }; | |
| const html = window.render(text, options); | |
| el.outerHTML = html; | |
| } | |
| }; | |
| </script> | |
| </head> | |
| <body> | |
| <div id="content"><div id="content-text"></div></div> | |
| </body> | |
| </html> | |
| """ | |
| tik_html = """ | |
| <!DOCTYPE html> | |
| <html> | |
| <head> | |
| <meta charset="UTF-8"> | |
| <meta name="viewport" content="width=device-width, initial-scale=1.0"> | |
| <title>Document</title> | |
| <link rel="stylesheet" type="text/css" href="https://tikzjax.com/v1/fonts.css"> | |
| <script src="https://tikzjax.com/v1/tikzjax.js"></script> | |
| </head> | |
| <body> | |
| <script type="text/tikz"> | |
| const text = | |
| </script> | |
| </body> | |
| </html>""" | |
| # print(tik_html) | |