reasoning benchmark
Frontier Science Leaderboard
Frontier Science is a benchmark of exceptionally challenging scientific reasoning problems spanning advanced natural-science domains, designed to test expert-level scientific understanding and multi-step reasoning.
Updated Aug 17, 2026
Models1
Model coverage1
MetricScore
EvidenceB
Frontier Science Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelMI | Score58.2% | Percentile100.0% | Participants1 | EvidenceC | Evaluated |
Frontier Science Highlights
The leading models and scores on this benchmark.
The Top AI Models for Frontier Science
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis frontier science AI model leaderboard uses descending score in the benchmark's original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.
- 01MIMAI-Code-1-FlashMicrosoft
- Score
- 58.2%
Strengths
- Ranks #1 of 1 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures Frontier Science, not total model capability
Selection summary
Best AI Models for Frontier Science
MAI-Code-1-Flash currently leads Frontier Science with 58.2%. It is the top model on this specific benchmark, while the best LLM for the broader task should also be checked against other benchmarks, price and runtime.
Use this leaderboard with the supporting benchmark results and coverage details above. A leaderboard position summarizes the selected ranking signal; it does not replace workload-specific testing.
What is Frontier Science?
What Frontier Science measures and how its scores work.
Frontier Science is a benchmark of exceptionally challenging scientific reasoning problems spanning advanced natural-science domains, designed to test expert-level scientific understanding and multi-step reasoning.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- Frontier Science
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- frontier-science|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about Frontier Science.
Which model scores highest on Frontier Science?
MAI-Code-1-Flash is currently ranked first with 58.2%.
What does Frontier Science measure?
Frontier Science is a benchmark of exceptionally challenging scientific reasoning problems spanning advanced natural-science domains, designed to test expert-level scientific understanding and multi-step reasoning.
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.