README.md

metadata

title: Unlimited OCR API
emoji: 🔍
colorFrom: blue
colorTo: purple
sdk: gradio
sdk_version: 5.44.1
app_file: app.py
pinned: false
license: mit
short_description: Baidu Unlimited-OCR — free document parsing API (ZeroGPU)
tags:
  - ocr
  - document-parsing
  - vision-language
  - baidu
  - zero-gpu
hardware: zero-a100

Unlimited OCR API

Free API for baidu/Unlimited-OCR — one-shot long-horizon document parsing.

Runs on ZeroGPU (free, no inference credits). ⚠️ Subject to queue/timeout.

Features

• 📄 Single image OCR (gundam/fast or base/accurate)
• 📚 Multi-image batch OCR
• 📑 PDF support (auto page extraction)
• 📡 REST API at /ocr and /ocr/pdf

API Usage

Python (gradio_client)

from gradio_client import Client

client = Client("Harry00/unlimited-ocr-api")

# Single image OCR
result = client.predict("doc.png", "gundam", "document parsing.", api_name="/handle_single")
print(result)

# PDF OCR
result = client.predict("scan.pdf", "Multi page parsing.", api_name="/handle_pdf")
print(result)

REST (curl)

# Single image
curl -X POST "https://harry00-unlimited-ocr-api.hf.space/ocr" \
  -F "image=@document.png" \
  -F "mode=gundam" \
  -F "prompt=document parsing."

# PDF
curl -X POST "https://harry00-unlimited-ocr-api.hf.space/ocr/pdf" \
  -F "pdf=@scan.pdf" \
  -F "prompt=Multi page parsing."

# Health check
curl https://harry00-unlimited-ocr-api.hf.space/health