Benchmarks · math
Riemann-Bench
Uses Surge AI's first-party extreme-math leaderboard score. Score: score. Collected by Jake Boggs' benchmark dashboard.
20 models · 20 results · data as of 2026-11-06 · surgehq.ai ↗
Leaderboard
GPT-5.6 Sol max
All results
20 rows| # | Model | Effort | Riemann-Bench |
|---|---|---|---|
| 1 | GPT-5.6 Sol | max | 83.2% |
| 2 | GPT-6 Astra | max | 81.6% |
| 3 | Opus 5.5 | max | 80.0% |
| 4 | Gemini 4 Carbon | high | 80.0% |
| 5 | GPT-6.1 Sol | max | 79.2% |
| 6 | Sonnet 5.5 | max | 76.8% |
| 7 | Gemini 4 Argon | high | 74.4% |
| 8 | Fable 5.1 | max | 71.2% |
| 9 | Muse Wonder Full | max | 63.2% |
| 10 | Kimi K3.1 | max | 60.8% |
| 11 | Muse Wonder (Max) | max | 59.2% |
| 12 | Grok 4.7 | xhigh | 57.6% |
| 13 | Gemini 3.8 Flash | high | 56.8% |
| 14 | Muse Spark 1.3 | max | 53.6% |
| 15 | Qwen 3.8 Max | max | 48.0% |
| 16 | DeepSeek V4.1 Flash 1017 | max | 38.4% |
| 17 | Kimi K3 | max | 36.8% |
| 18 | Hy4 preview | high | 24.0% |
| 19 | GLM 5.3 | max | 20.8% |
| 20 | Mistral Large 4 | high | 20.0% |