OpenGenerativeAI/llm-colosseum

Benchmark LLMs by fighting in Street Fighter 3! The new way to evaluate the quality of an LLM

47
/ 100
Emerging

This project helps evaluate and compare different Large Language Models (LLMs) by having them compete in the video game Street Fighter III. You provide the LLMs you want to test, and the system shows which one performs best based on criteria like speed, smart decision-making, and adaptability. It's designed for researchers or practitioners who need to assess LLM capabilities beyond standard benchmarks.

1,467 stars. No commits in the last 6 months.

Use this if you need a dynamic, real-time method to benchmark the decision-making, speed, and strategic capabilities of different LLMs in a complex, adversarial environment.

Not ideal if you are solely interested in traditional linguistic or factual accuracy benchmarks for LLMs, as this focuses on game-play performance.

LLM-evaluation AI-benchmarking strategic-AI real-time-decisioning agent-testing
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 21 / 25

How are scores calculated?

Stars

1,467

Forks

178

Language

Jupyter Notebook

License

MIT

Last pushed

Mar 21, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/OpenGenerativeAI/llm-colosseum"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.