Test LangChain pipelines against multiple LLM providers

This integration example demonstrates how to run Python LangChain Expression Language (LCEL) chains within Promptfoo evaluation workflows. The example compares GPT models with a math-focused LangChain prompt-and-output-parser pipeline.

Use this integration when you want to run Python LangChain LCEL chains with Promptfoo's evaluation framework. This example is specifically designed for Python LangChain implementations.

Do not use this if you're working with LangChain JavaScript/TypeScript implementations, as this example specifically targets Python LangChain. Avoid this approach if you don't need the complexity of LCEL chains and can accomplish your goals with direct LLM API calls.

You provide a Python environment (3.10 or newer), OpenAI API credentials, and your LangChain LCEL chain definitions. The example includes requirements for dependencies and configuration for running evaluations.

You receive a working example that runs LangChain chains within Promptfoo.

This example integrates Python LangChain Expression Language (LCEL) chains with Promptfoo's evaluation framework. It uses OpenAI models as providers. The setup requires Python 3.10+ with a virtual environment and pip-managed dependencies.

To get started, initialize the example:

npx promptfoo@latest init --example integration-langchain
cd integration-langchain

Then set up the Python environment:

python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

Configure your OpenAI credentials:

export OPENAI_API_KEY=your-api-key

Run the evaluation:

px promptfoo eval

This integration example is for users who want to run Python LangChain LCEL chains with Promptfoo.