GeneBench-Pro Artificial Analysis Subset
Science · 2026-06-30
GeneBench-Pro Artificial Analysis subset reports pass rates on the 50 held-out problems provided to Artificial Analysis for independent third-party reporting.
Top models (higher is better)
| Model | Score |
|---|---|
| GPT-5.6 Sol Pro | 18.0 |
| GPT-5.6 Terra Pro | 17.6 |
| GPT-5.6 Luna Pro | 16.4 |
| GPT-5.6 Sol | 16.3 |
| GPT-5.6 Terra | 16.2 |
| GPT-5.6 Luna | 9.6 |
| GPT-5.4 Pro | 9.2 |
| GPT-5.5 Pro | 9.2 |
| Gemini 3.5 Flash | 6.0 |
| Opus 4.8 | 4.8 |
| GPT-5.4 | 4.7 |
| GPT-5.5 | 4.6 |
| GLM-5.2 | 3.4 |
| Qwen3.7-Max | 3.2 |
| GPT-5.2 | 3.2 |
| GPT-5.2 Pro | 2.4 |
| DeepSeek-V4-Pro | 2.3 |
| MiMo-V2.5-Pro | 1.8 |
| Kimi K2.7 Code | 1.8 |
| Qwen3.7-Plus | 1.8 |
| DeepSeek-V4-Flash | 1.2 |
| Gemini 3.1 Pro Preview | 1.0 |
| MiniMax M3 | 0.7 |
| GLM-5.1 | 0.7 |
| Grok 4.3 | 0.6 |
| Kimi K2.6 | 0.6 |
| Hy3 preview | 0.4 |
| MiMo-V2.5 | 0.4 |
| MiniMax M2.7 | 0.0 |