TestsBenchmarks · Math
FrontierMath Tier 4
Accuracy on the hardest FrontierMath tier of research-level problems (Epoch-run, v2).
% solvedHigher is betterToo easy now: the top models all score near perfectSaturatedThe test's websiteOfficial page ↗
Measured by independent testersIndependent Reported by the makerVendor-reported
| Thinking levelSetting | |||||||
|---|---|---|---|---|---|---|---|
| 1 | GPT-6.1 Sol | OpenAI | 100.0% | Maxgpt-6.1-sol_max | Independent testIndependentEpoch AI ↗ | — | 29 Sep 2026 |
| 2 | GPT-6 Astra | OpenAI | 97.6% | Highgpt-6-astra_high | Independent testIndependentEpoch AI ↗ | — | 30 Aug 2026 |
| 3 | Claude Opus 5.5 | Anthropic | 95.0% | Maxclaude-opus-5-5_max | Independent testIndependentEpoch AI ↗ | — | 22 Sep 2026 |
| 4 | Claude Fable 5 | Anthropic | 90.2% | Maxclaude-fable-5_max | Independent testIndependentEpoch AI ↗ | — | 9 Jun 2026 |
| 5 | GPT-6 Sol | OpenAI | 90.0% | Maxgpt-6-sol_max | Independent testIndependentEpoch AI ↗ | — | 22 Sep 2026 |
| 6 | Claude Fable 5.1 | Anthropic | 87.8% | Maxclaude-fable-5-1_max | Independent testIndependentEpoch AI ↗ | — | 1 Sep 2026 |
| 7 | GPT-5.6 Sol | OpenAI | 83.0% | Defaultbest score across efforts (OpenAI table: "maximum at any effort") | Maker's own figureVendor-reportedOpenAI ↗ | — | 3 Sep 2026 |
| 8 | Claude Sonnet 5.5 | Anthropic | 80.5% | Maxclaude-sonnet-5-5_max | Independent testIndependentEpoch AI ↗ | — | 29 Sep 2026 |
| 9 | GPT-5.6 Sol Pro | OpenAI | 80.5% | Maxgpt-5.6-sol_promax | Independent testIndependentEpoch AI ↗ | — | 9 Jul 2026 |
| 10 | GPT-5.5 Pro | OpenAI | 78.0% | Extra highgpt-5.5-pro_xhigh | Independent testIndependentEpoch AI ↗ | — | 12 Jun 2026 |
| 11 | Claude Opus 5 | Anthropic | 73.2% | Maxclaude-opus-5_max | Independent testIndependentEpoch AI ↗ | — | 24 Jul 2026 |
| 12 | GPT-5.5 | OpenAI | 72.5% | Extra highgpt-5.5_xhigh | Independent testIndependentEpoch AI ↗ | — | 11 Jun 2026 |
| 13 | GPT-5.6 Terra | OpenAI | 70.7% | Maxgpt-5.6-terra_max | Independent testIndependentEpoch AI ↗ | — | 9 Jul 2026 |
| 14 | GPT-5.6 Luna | OpenAI | 61.0% | Maxgpt-5.6-luna_max | Independent testIndependentEpoch AI ↗ | — | 9 Jul 2026 |
| 15 | GPT-5.4 Pro | OpenAI | 58.5% | Extra highgpt-5.4-pro-2026-03-05_xhigh | Independent testIndependentEpoch AI ↗ | — | 13 Jun 2026 |
| 16 | GPT-6 Luna | OpenAI | 56.1% | Maxgpt-6-luna_max | Independent testIndependentEpoch AI ↗ | — | 22 Sep 2026 |
| 17 | Claude Opus 4.8 | Anthropic | 56.1% | Maxclaude-opus-4-8_max | Independent testIndependentEpoch AI ↗ | — | 10 Jun 2026 |
| 18 | GPT-5.4 | OpenAI | 27.1% | Extra highreasoning effort=xhigh | Maker's own figureVendor-reportedOpenAI ↗ | — | 23 Apr 2026 |