FrontierMath Tier 1-3 (v2)Browse 296

FrontierMath Tier 1-3 (v2)

Can the model solve difficult problems that require more than factual recall?

Version not specifiedExact reported variant
Reasoning7 ranked models7 reported values1 reportsHigher is better

Top rankings

One row per model, using its best reported score across effort settings.

7
RankModelBest scoreBest reported setting
1GPT-5.6 SolOpenAI89.0%Reported configurationOpenAI report1 value · 1 reportJul 9, 2026 · source
2Claude Fable 5Anthropic87.0%Reported configurationOpenAI report1 value · 1 reportJul 9, 2026 · source
3GPT-5.5OpenAI85.3%Reported configurationOpenAI report1 value · 1 reportJul 9, 2026 · source
4GPT-5.6 TerraOpenAI84.9%Reported configurationOpenAI report1 value · 1 reportJul 9, 2026 · source
5Claude Opus 4.8Anthropic80.0%Reported configurationOpenAI report1 value · 1 reportJul 9, 2026 · source
6GPT-5.6 LunaOpenAI78.6%Reported configurationOpenAI report1 value · 1 reportJul 9, 2026 · source
7Gemini 3.1 Pro PreviewGoogle59.6%Reported configurationOpenAI report1 value · 1 reportJul 9, 2026 · source

Effort curve

Every sourced cost-linked effort value for this exact version. Lines connect complete sweeps only.

0
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