Skip to content

openai/parameter-golf

Official

Train the smallest LM you can that fits in 16MB. Best model wins!

5,072 stars3,350 forksPythonUpdated May 2026
✅ Reviewed by My AI Guide

Our Verdict

OpenAI's Parameter Golf challenge to train the smallest language model fitting in 16MB: best model wins. For ML researchers pushing model compression limits. Unique competitive leaderboard ranks top tiny LMs by performance.

Frequently Asked Questions

What is openai/parameter-golf?

It is a competition where participants train the smallest possible language model that fits within 16MB, and the best-performing model wins. Implemented in Python, it drives innovation in ultra-efficient AI models.

What is parameter-golf?

Train the smallest LM you can that fits in 16MB. Best model wins!

How do I install parameter-golf?

Visit the GitHub repository at https://github.com/openai/parameter-golf for installation instructions.

What license does parameter-golf use?

parameter-golf uses the MIT license.

What are alternatives to parameter-golf?

Search My AI Guide for similar tools in this category.

🔒

Open source & community-verified

MIT licensed: free to use in any project, no strings attached. 5,072 developers have starred this, meaning the community has reviewed and trusted it.

Reviewed by My AI Guide for relevance, quality, and active maintenance before listing.