Skip to main content

​
Prerequisites

LangExtract requires Python 3.10 or higher.
1

Choose Your Installation Method

Select the installation method that best fits your needs:
  • PyPI: Recommended for most users
  • Virtual Environment: Best for isolated development
  • From Source: For contributors and advanced users
  • Docker: For containerized deployments
2

Install LangExtract

pip install langextract
Installing with -e puts the package in development mode, allowing you to modify the code without reinstalling.
3

Install Optional Dependencies

LangExtract supports optional providers and tools:
OpenAI Support
pip install langextract[openai]
All Optional Dependencies
pip install langextract[all]
Development Tools
pip install langextract[dev]
Testing Framework
pip install langextract[test]
Jupyter Notebook
pip install langextract[notebook]
4

Set Up API Key

Cloud-hosted models like Gemini require an API key. Local models with Ollama don’t need one.

​
Get Your API Key

Choose your provider:

​
Configure Your API Key

export LANGEXTRACT_API_KEY="your-api-key-here"
Never commit API keys to version control. Always use environment variables or .env files.
5

Verify Installation

Test your installation:
import langextract as lx
print(lx.__version__)  # Should print: 1.1.1

​
Core Dependencies

LangExtract requires the following packages:
  • google-genai (>=1.39.0) - Gemini model support
  • aiohttp (>=3.8.0) - Async HTTP requests
  • pydantic (>=1.8.0) - Data validation
  • pandas (>=1.3.0) - Data processing
  • python-dotenv (>=0.19.0) - Environment variable management
  • tqdm (>=4.64.0) - Progress bars
See the full dependency list in pyproject.toml.

​
Using Local LLMs with Ollama

Run LangExtract without API keys using Ollama:
1

Install Ollama

Download from ollama.com
2

Pull a Model

ollama pull gemma2:2b
3

Start Ollama Server

ollama serve
4

Use in LangExtract

import langextract as lx

result = lx.extract(
    text_or_documents=input_text,
    prompt_description=prompt,
    examples=examples,
    model_id="gemma2:2b",
    model_url="http://localhost:11434",
    fence_output=False,
    use_schema_constraints=False
)

​
Troubleshooting

​
ImportError

If you encounter import errors, ensure you’re using Python 3.10 or higher: 
python --version

​
API Key Issues

If your API key isn’t being recognized:
  1. Check that LANGEXTRACT_API_KEY is set in your environment
  2. Verify the .env file is in your working directory
  3. Ensure you’re loading the .env file in your code if needed

​
Docker Issues

If Docker build fails, ensure Docker is installed and running: 
docker --version

​
Next Steps

Now that LangExtract is installed, head to the Quick Start guide to run your first extraction.

Build docs developers (and LLMs) love

Get started for freeTalk to us