MLE-bench-30 selected subset; <50GB data and <10h reference-task filter; bronze pass@1; agent has a GPU and Kaggle data. Generic environment description allows 24h and some 100h experiments; exact per-model duration unspecified.
GPT-5.1 Codex-Max · MLE-Bench · 30 competition subset selected from tasks with <50GB data and <10h reference duration · Scores and configurations
GPT-5.1 Codex-Max record for MLE-Bench · 30 competition subset selected from tasks with <50GB data and <10h reference duration. 1 score record for this release, including 1 exact value (may include historical records). Includes 1 source location.
MLE-bench-30 selected subset; <50GB data and <10h reference-task filter; bronze pass@1; agent has a GPU and Kaggle data. Generic environment description allows 24h and some 100h experiments; exact per-model duration unspecified.