Flight search assistant with tool calling

View Source Code

Browse the complete example on GitHub

This project demonstrates a Python CLI that leverages the LFM2.5-1.2B-Thinking model to help users find and book plane tickets through multi-step reasoning and tool calling.

Quickstart

1. Verify uv installation

Check that you have uv available on your system:

uv --version

If this fails, install uv following these instructions.

2. Clone and build the project

Clone the repository and install dependencies:

git clone https://github.com/Liquid4All/cookbook.git
cd cookbook/examples/flight-search-assistant
uv sync

3. Run example queries

The assistant accepts natural language requests through the command-line interface:

Simple flight availability search:

uv run flight_search.py --query "What flights are available from New York to Paris on 2026-01-19?"

Direct booking request:

uv run flight_search.py --query "Book flight AA495 for 2026-02-04"

Multi-step workflow (search then book):

uv run flight_search.py --query "Book the cheapest flight from Barcelona to Belgrade on 2026-01-31"

Complex reasoning task:

uv run flight_search.py --query "Book the cheapest flight from Barcelona to a US city on the East Coast that is not NYC on 2026-02-14"

The implementation uses mocked flight data for demonstration purposes, though it supports integration with real APIs like Amadeus, Skyscanner, or Kiwi for production deployment.

Future Development

The roadmap includes:

• Adding evaluation datasets
• Implementing performance improvements through GRPO fine-tuning with verifiable rewards

Need help?

Join our Discord

Connect with the community and ask questions about this example.

Edit this page