semiautomatix/deepseek-ocr-token-test

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

Running Locally

DeepSeek-OCR relies on CUDA for inference. Make sure you are using a GPU-enabled runtime with CUDA 12-compatible drivers.

1. 1.Create and activate your GPU Python environment.
2. 2.Install dependencies:

   pip install -r requirements.txt

1. 1.Launch the demo:

   python app.py

1. 1.Open http://localhost:7860/ in your browser, upload an image, and click Run OCR to view the extracted text and token usage.

Hugging Face Space

Deploy the Space on GPU hardware (for ZeroGPU, the first request will provision a temporary GPU). The Gradio Blocks UI exposes an image upload, text output, and token usage panel backed by the deepseek-ai/DeepSeek-OCR model.