Langextract Web

🚀 LangExtract Web

Extract structured information from text using LLMs. A web UI + API + MCP wrapper for Google's LangExtract library.

🚀 Quick Start

Docker (Recommended)

docker run -d -p 8600:8600 \
  -e LANGEXTRACT_API_KEY=your-gemini-api-key \
  neosun/langextract:latest

Open http://localhost:8600

Docker Compose

services:
  langextract:
    image: neosun/langextract:latest
    ports:
      - "8600:8600"
    environment:
      - LANGEXTRACT_API_KEY=${LANGEXTRACT_API_KEY}
      - OPENAI_API_KEY=${OPENAI_API_KEY}
      - OLLAMA_HOST=http://host.docker.internal:11434
    volumes:
      - /tmp/langextract:/tmp/langextract
    extra_hosts:
      - "host.docker.internal:host-gateway"

docker compose up -d

✨ Features

• 🎯 Precise Source Grounding — Every extraction maps back to exact text positions
• 📚 Few-shot Learning — Define extraction tasks with just a few examples, no fine-tuning needed
• 📄 Long Document Optimization — Chunking + multi-pass extraction for high recall
• 🌐 Web UI — Modern, responsive interface with dark mode and i18n support
• 🔌 REST API — Full Swagger documentation at /docs
• 🤖 MCP Support — Model Context Protocol for AI assistant integration
• 🔧 LiteLLM Compatible — Use any LLM provider (Gemini, OpenAI, Claude, Ollama, etc.)
• 📁 File Upload — Drag & drop files or fetch from URL

📦 Installation

From Source

git clone https://github.com/neosun100/langextract-web.git
cd langextract-web

# Install dependencies
pip install -e .
pip install flask flask-cors flasgger gunicorn

# Run
python app.py

Environment Variables

💻 Usage Examples

Web UI

1. Open http://localhost:8600
2. Enter text or drag & drop a file / paste URL
3. Define extraction prompt and few-shot examples
4. Select model and configure parameters
5. Click "Extract" and view results with visualization

REST API

# Health check
curl http://localhost:8600/health

# Extract
curl -X POST http://localhost:8600/api/extract \
  -H "Content-Type: application/json" \
  -d '{
    "text": "Lady Juliet gazed at the stars, her heart aching for Romeo.",
    "prompt": "Extract characters and emotions",
    "examples": [{
      "text": "ROMEO spoke softly",
      "extractions": [{"extraction_class": "character", "extraction_text": "ROMEO"}]
    }],
    "model_id": "gemini-2.5-flash"
  }'

Full API documentation: http://localhost:8600/docs

MCP Integration

{
  "mcpServers": {
    "langextract": {
      "command": "docker",
      "args": ["exec", "-i", "langextract", "python", "mcp_server.py"]
    }
  }
}

🔧 Technical Details

Extraction Parameters

Supported Models

🏗️ Tech Stack

• Backend: Flask, Gunicorn
• Core: LangExtract by Google
• LLM: Google Gemini, OpenAI, Anthropic, Ollama
• Container: Docker

📝 Changelog

v1.2.0

• ✨ File drag & drop upload
• ✨ URL content fetching
• ✨ Custom model support (LiteLLM)
• ✨ Full parameter exposure in UI
• ✨ Project introduction section

v1.0.0

• 🎉 Initial release with Web UI + API + MCP

🤝 Contributing

Contributions welcome! Please read the Contributing Guide.

📄 License

Apache 2.0 - See LICENSE

Based on LangExtract by Google.

⭐ Star History

📱 Follow Us