FrontierMath Tier 4 (v2)Browse 284 families
FrontierMath Tier 4 (v2)
Can the model solve difficult problems that require more than factual recall?
Version / measure
- Domain
- Reasoning
- Ranked models
- 6
- Reported values
- 6
- Reports
- 1 · lab
- Direction
- Higher is better
- Status
- Version not specified
Rankings
Best reported score per model across effort settings
| # | Model | Best setting | Score |
|---|---|---|---|
| 1 | Claude Fable 5Anthropic | reportedcost n/a1 value · Jul 9 · source | 87.8% |
| 2 | GPT-5.6 SolOpenAI | reportedcost n/a1 value · Jul 9 · source | 83.0% |
| 3 | GPT-5.5OpenAI | reportedcost n/a1 value · Jul 9 · source | 72.5% |
| 4 | GPT-5.6 TerraOpenAI | reportedcost n/a1 value · Jul 9 · source | 68.3% |
| 5 | GPT-5.6 LunaOpenAI | reportedcost n/a1 value · Jul 9 · source | 58.5% |
| 6 | Claude Opus 4.8Anthropic | reportedcost n/a1 value · Jul 9 · source | 56.1% |
Effort curve
Score vs. reported cost. Lines connect complete sweeps only.
No cost-linked effort sweep for this version.
Definition and comparison boundarypublic methodology
Can the model solve difficult problems that require more than factual recall?
Multi-step reasoning on academic, mathematical, or abstract problems. Higher is better. The value is the percentage reported in this lab's table.
Harness, tools, prompts, attempt count, and benchmark version can materially change the result. Compare within one reporting context.
Method / source ↗