math benchmark
FrontierMath Tier 4 (v2)
FrontierMath Tier 4 subset from the v2 evaluation release.
Updated Aug 7, 2026
Published models3
Registry coverage3
MetricScore
EvidenceB
FrontierMath Tier 4 (v2) leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | OP | 83.0% | 100.0% | 3 | C | |
| 2 | OP | 68.3% | 50.0% | 3 | C | |
| 3 | OP | 58.5% | 0.0% | 3 | C |
Score distribution
Top published rows on the benchmark's original scale.
FrontierMath Tier 4 (v2) highlights
The top published results on this benchmark's own scale.
What is FrontierMath Tier 4 (v2)?
Definition and scoring fields from the benchmark registry.
FrontierMath Tier 4 subset from the v2 evaluation release.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- FrontierMath Tier 4 (v2)
- Modality
- text
- Primary category
- math
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- frontiermath-tier-4-v2|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about FrontierMath Tier 4 (v2).
Which model scores highest on FrontierMath Tier 4 (v2)?
GPT-5.6 Sol is currently ranked first with 83.0%.
What does FrontierMath Tier 4 (v2) measure?
FrontierMath Tier 4 subset from the v2 evaluation release.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
3 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.