Benchmark
FrontierMath Tier 1-3 (v2)
Math benchmarks measure how well models solve maths problems. Scores are percentages. Higher is better.
- Models measured
- 11
- Scores recorded
- 13
- Sources
- 1
- IQ weight
- 0.75
Score distribution
- 65.2%
- Mean
- 59.6%
- Median
- 18.3%
- Std dev
- 40.7%
- 10th percentile
- 87%
- 90th percentile
- 40.3%
- Min
This benchmark has a weight of 0.75 for chat models, measured over 11 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.6 Solv2· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 89% |
| 2 | Claude Fable 5v2· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 87% |
| 3 | GPT 5.5v2· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 85.3% |
| 4 | GPT-5.6 Terrav2· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 84.9% |
| 5 | Claude Opus 4.8v2· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 80% |
| 6 | GPT-5.6 Lunav2· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 78.6% |
| 7 | Gemini 3.1 Pro Previewv2· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 59.6% |
| 8 | GPT-5.5 Prov2 · xhigh· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 52.4% |
| 9 | GPT 5.5v2 · xhigh· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 51.7% |
| 10 | GPT-5.4 ProOpenAI GPT-5.4 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 50% |
| 11 | GPT 5.4OpenAI GPT-5.4 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 47.6% |
| 12 | GPT 5.2OpenAI GPT-5.4 launch table· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 40.7% |
| 13 | GPT 5.2Python · xhigh· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 40.3% |