GeneBench-Pro Public Release Subset
Science · 2026-06-30
GeneBench-Pro public release subset reports pass rates on the 10 final public case-study problems from the 129-problem GeneBench-Pro suite.
Top models (higher is better)
| Model | Score |
|---|---|
| GPT-5.6 Terra Pro | 36.0 |
| GPT-5.6 Sol Pro | 34.0 |
| GPT-5.6 Sol | 24.0 |
| GPT-5.6 Luna Pro | 22.0 |
| GPT-5.4 Pro | 20.0 |
| Opus 4.8 | 18.0 |
| GPT-5.5 Pro | 14.0 |
| GPT-5.6 Luna | 9.0 |
| GLM-5.2 | 8.9 |
| GPT-5.4 | 8.0 |
| GPT-5.5 | 8.0 |
| GPT-5.6 Terra | 8.0 |
| GPT-5.2 Pro | 8.0 |
| Kimi K2.6 | 7.7 |
| Gemini 3.1 Pro Preview | 7.0 |
| Qwen3.7-Max | 6.4 |
| MiMo-V2.5-Pro | 6.0 |
| DeepSeek-V4-Flash | 5.6 |
| Gemini 3.5 Flash | 5.0 |
| Qwen3.7-Plus | 4.4 |
| MiniMax M3 | 4.0 |
| GPT-5.2 | 4.0 |
| MiMo-V2.5 | 3.0 |
| DeepSeek-V4-Pro | 2.0 |
| Hy3 preview | 1.0 |
| MiniMax M2.7 | 0.0 |
| GLM-5.1 | 0.0 |
| Grok 4.3 | 0.0 |
| Kimi K2.7 Code | 0.0 |