New: GPT-6.2 benchmark report is here.Read the report ↗

Best LLMs on MLE-Bench Revised

MLE-Bench Revised

Trains and refines machine-learning models on competition data and ranks hidden-test performance against reference solutions.

Updated: Data as of:
Limited reportingSaturation: not applicable
Official benchmark
MLE-Bench Revised results. Each row identifies its benchmark version and evaluation track. Scores from different versions are not directly comparable. Best effort from the latest stored report date per model and version.
Model / creatorVersionReported score
GPT-6.2 AstraOpenAI · ProprietaryGPT-6.2 system card96.06 percentile
GPT-6.2 SolOpenAI · ProprietaryGPT-6.2 system card91.38 percentile
View all 2 published reports