agents benchmark
MLE-Bench
MLE-Bench evaluates AI agents on machine learning engineering tasks by measuring their performance on Kaggle competitions.
Updated Aug 11, 2026
Models2
Model coverage2
MetricScore
EvidenceB
MLE-Bench Ranking
Higher score ranks better on this benchmark.
| 01 | GO | 63.9% | 100.0% | 2 | C | |
| 02 | GO | 39.2% | 0.0% | 2 | C |
MLE-Bench Score Distribution
A closer view of the leading scores on this benchmark.
MLE-Bench Highlights
The leading models and scores on this benchmark.
What is MLE-Bench?
What MLE-Bench measures and how its scores work.
MLE-Bench evaluates AI agents on machine learning engineering tasks by measuring their performance on Kaggle competitions.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- MLE-Bench
- Modality
- text
- Primary category
- agents
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- mle-bench|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about MLE-Bench.
Which model scores highest on MLE-Bench?
Gemini 3.6 Flash is currently ranked first with 63.9%.
What does MLE-Bench measure?
MLE-Bench evaluates AI agents on machine learning engineering tasks by measuring their performance on Kaggle competitions.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
2 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.