KellyBench
KellyBench is a benchmark that tests an agents' ability to make machine learning models for predicting football matches and betting against market odds.
- Domain
- rl-env
- License
- unknown
- Published
- Jan 2026
Cite
Notes
Only stored in your browser.
Top score 92063 by GPT-5.4 - 5 models reporting (2 frontier)
Score history
3Top models
5
FAQ
- What is KellyBench?
- KellyBench is a benchmark that tests an agents' ability to make machine learning models for predicting football matches and betting against market odds.
- What is the current top score on KellyBench?
- The top reported score is 92063 by GPT-5.4, across 5 models reporting (2 from frontier labs).
- What license is KellyBench under?
- KellyBench is available under unknown.