Updated daily at 05:00 UTC
Benchmark
FrontierMath Tier 4
Measures: 43 exceptionally hard, research-level unpublished math problems (the FrontierMath expansion tier) with automatically verifiable answers; 1 point per correct answer on the private set.
- Maintainer
- Epoch AI (OpenAI-funded; conflict of interest disclosed)
- Introduced
- 2025
- Status
- Active
- Saturation
- Saturated
- Typical provenance
- Independent
- In scoring set
- No
Leaderboard
Independent and third-party results only.
| Model | Accuracy | Provenance | Source | Evaluated |
|---|---|---|---|---|
| GPT-6 Astra OpenAI · High effort | 97.6% | Independent | Epoch AI | 30 Aug 2026 |
| GPT-6 Astra OpenAI · Xhigh effort | 97.6% | Independent | Epoch AI | 30 Aug 2026 |
| GPT-6 Astra OpenAI · Max effort | 97.6% | Independent | Epoch AI | 30 Aug 2026 |
| GPT-6 Astra OpenAI · Medium effort | 97.6% | Independent | Epoch AI | 30 Aug 2026 |
| Claude Opus 5.5 Anthropic · Max effort | 95.0% | Independent | Epoch AI | 22 Sep 2026 |
| Claude Fable 5 Anthropic · Max effort | 90.2% | Independent | Epoch AI | 9 Jun 2026 |
| Claude Fable 5.1 Anthropic · Max effort | 87.8% | Independent | Epoch AI | 1 Sep 2026 |
| GPT-6 Astra OpenAI · Low effort | 87.8% | Independent | Epoch AI | 30 Aug 2026 |
| GPT-5.6 Sol OpenAI · Max effort | 82.9% | Independent | Epoch AI | 9 Jul 2026 |
| GPT-6 Astra OpenAI · None effort | 82.9% | Independent | Epoch AI | 30 Aug 2026 |
| GPT-5.6 Sol OpenAI · Promax effort | 80.5% | Independent | Epoch AI | 9 Jul 2026 |
| GPT-5.5 Pro OpenAI · Xhigh effort | 78.0% | Independent | Epoch AI | 12 Jun 2026 |
| Claude Opus 5 Anthropic · Max effort | 73.2% | Independent | Epoch AI | 24 Jul 2026 |
| GPT-5.5 OpenAI · Xhigh effort | 72.5% | Independent | Epoch AI | 11 Jun 2026 |
| GPT-5.6 Terra OpenAI · Max effort | 70.7% | Independent | Epoch AI | 9 Jul 2026 |
| GPT-5.6 Luna OpenAI · Max effort | 61.0% | Independent | Epoch AI | 9 Jul 2026 |
| GPT-5.4 Pro OpenAI · Xhigh effort | 58.5% | Independent | Epoch AI | 13 Jun 2026 |
| Claude Opus 4.8 Anthropic · Max effort | 56.1% | Independent | Epoch AI | 10 Jun 2026 |
| GPT-5.4 OpenAI · Xhigh effort | 49.0% | Independent | Epoch AI | 11 Jun 2026 |
| Muse Spark 1.3 Meta · Max effort | 46.3% | Independent | Epoch AI | 18 Sep 2026 |
| Qwen3.8-Max Alibaba Group · Xhigh effort | 46.3% | Independent | Epoch AI | 4 Aug 2026 |
| GPT-5.2 Pro OpenAI · Xhigh effort | 46.0% | Independent | Epoch AI | 13 Jun 2026 |
| Muse Spark 1.3 Meta · Xhigh effort | 41.5% | Independent | Epoch AI | 16 Sep 2026 |
| Kimi K3 Moonshot AI · Max effort | 39.0% | Independent | Epoch AI | 17 Jul 2026 |
| Gemini 3.7 Flash Google DeepMind · High effort | 36.6% | Independent | Epoch AI | 14 Aug 2026 |
| Qwen3.7-Max | 34.1% | Independent | Epoch AI | 13 Jun 2026 |
| Qwen3.8-Max Alibaba Group · Xhigh effort | 34.1% | Independent | Epoch AI | 2 Sep 2026 |
| Claude Opus 4.7 Anthropic · Max effort | 31.7% | Independent | Epoch AI | 10 Jun 2026 |
| Grok 4.6 SpaceXAI (xAI) · Xhigh effort | 31.7% | Independent | Epoch AI | 14 Aug 2026 |
| GPT-5.2 OpenAI · Xhigh effort | 31.7% | Independent | Epoch AI | 11 Jun 2026 |
| Claude Sonnet 5 Anthropic · Max effort | 29.3% | Independent | Epoch AI | 30 Jun 2026 |
| GLM-5.2 Z.ai (Zhipu AI) · Max effort | 29.3% | Independent | Epoch AI | 19 Jun 2026 |
| GLM-5.3 Z.ai (Zhipu AI) · Max effort | 29.3% | Independent | Epoch AI | 25 Aug 2026 |
| Claude Opus 4.6 Anthropic · Max effort | 26.8% | Independent | Epoch AI | 11 Jun 2026 |
| DeepSeek-V4-Pro DeepSeek · Max effort | 26.8% | Independent | Epoch AI | 19 Aug 2026 |
| Gemini 3.1 Pro | 26.8% | Independent | Epoch AI | 11 Jun 2026 |
| Gemini 3.5 Flash Google DeepMind · High effort | 26.8% | Independent | Epoch AI | 10 Jun 2026 |
| Kimi K2.6 | 25.6% | Independent | Epoch AI | 10 Jun 2026 |
| DeepSeek-V4-Flash DeepSeek · Max effort | 24.4% | Independent | Epoch AI | 2 Aug 2026 |
| Grok 4.5 SpaceXAI (xAI) · High effort | 24.4% | Independent | Epoch AI | 9 Jul 2026 |
| Gemini 3.6 Flash Google DeepMind · High effort | 22.0% | Independent | Epoch AI | 2 Aug 2026 |
| Gemini 3.8 Flash Google DeepMind · High effort | 22.0% | Independent | Epoch AI | 2 Sep 2026 |
| GPT-5 OpenAI · High effort | 22.0% | Independent | Epoch AI | 11 Jun 2026 |
| GPT-5 Pro OpenAI · High effort | 19.5% | Independent | Epoch AI | 12 Jun 2026 |
| Gemini 3 Flash | 17.1% | Independent | Epoch AI | 11 Jun 2026 |
| GLM-5.3-Flash Z.ai (Zhipu AI) · Max effort | 17.1% | Independent | Epoch AI | 27 Aug 2026 |
| Grok 4.20 | 17.1% | Independent | Epoch AI | 13 Jul 2026 |
| Inkling Small Thinking Machines Lab · Xhigh effort | 17.1% | Independent | Epoch AI | 14 Aug 2026 |
| Grok 4.3 SpaceXAI (xAI) · High effort | 14.6% | Independent | Epoch AI | 17 Jun 2026 |
| GPT-5 mini OpenAI · High effort | 12.2% | Independent | Epoch AI | 12 Jun 2026 |
| GPT-5.4 nano OpenAI · High effort | 12.2% | Independent | Epoch AI | 12 Jun 2026 |
| Kimi K2.7 Code | 12.2% | Independent | Epoch AI | 13 Jun 2026 |
| GPT-5.4 mini OpenAI · Xhigh effort | 9.8% | Independent | Epoch AI | 12 Jun 2026 |
| Claude Opus 4.5 Anthropic · 32K effort | 4.9% | Independent | Epoch AI | 11 Jun 2026 |
| Inkling Thinking Machines Lab · Xhigh effort | 4.9% | Independent | Epoch AI | 6 Aug 2026 |
| o4-mini OpenAI · High effort | 4.9% | Independent | Epoch AI | 11 Jun 2026 |
| Claude Opus 4.1 Anthropic · 32K effort | 2.4% | Independent | Epoch AI | 11 Jun 2026 |
| Claude Sonnet 4.5 Anthropic · 32K effort | 2.4% | Independent | Epoch AI | 11 Jun 2026 |
| DeepSeek-V4-Pro DeepSeek · Max effort | 2.4% | Independent | Epoch AI | 17 Jun 2026 |
| GPT-5 nano OpenAI · High effort | 2.4% | Independent | Epoch AI | 12 Jun 2026 |
| Gemini 2.5 Pro | 0.0% | Independent | Epoch AI | 11 Jun 2026 |
| Gemini 3.5 Flash-Lite Google DeepMind · High effort | 0.0% | Independent | Epoch AI | 2 Aug 2026 |
| o3-mini OpenAI · High effort | 0.0% | Independent | Epoch AI | 11 Jun 2026 |
- Epoch AI, 'Capabilities & benchmarking', epoch.ai/benchmarks, CC BY 4.0.