agents benchmark
MLE-Bench Lite
MLE-Bench Lite evaluates AI agents on machine learning engineering tasks, testing their ability to build, train, and optimize ML models for Kaggle-style competitions in a lightweight evaluation format.
Updated Aug 11, 2026
Models1
Model coverage1
MetricScore
EvidenceB
MLE-Bench Lite Ranking
Higher score ranks better on this benchmark.
| 1 | MI | 66.6% | 100.0% | 1 | C |
MLE-Bench Lite Highlights
The leading models and scores on this benchmark.
What is MLE-Bench Lite?
What MLE-Bench Lite measures and how its scores work.
MLE-Bench Lite evaluates AI agents on machine learning engineering tasks, testing their ability to build, train, and optimize ML models for Kaggle-style competitions in a lightweight evaluation format.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- MLE-Bench Lite
- Modality
- text
- Primary category
- agents
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- mle-bench-lite|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about MLE-Bench Lite.
Which model scores highest on MLE-Bench Lite?
MiniMax M2.7 is currently ranked first with 66.6%.
What does MLE-Bench Lite measure?
MLE-Bench Lite evaluates AI agents on machine learning engineering tasks, testing their ability to build, train, and optimize ML models for Kaggle-style competitions in a lightweight evaluation format.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
1 model results are currently shown.
Does this benchmark affect the overall score?
No. This benchmark is shown for reference but does not contribute to the overall score.