FrontierCode 1.0 Diamond - Pass Rate
Code · 2026-06-08
FrontierCode 1.0 Diamond pass rate on the original benchmark's 50 hardest production-code tasks. A trial passes only when it satisfies every blocker criterion, approximating whether a repository maintainer would merge the patch. Cognition deprecated Diamond in FrontierCode 1.1.
Top models (higher is better)
| Model | Score |
|---|---|
| Opus 4.8 | 14.5 |
| GPT-5.5 | 7.2 |
| Opus 4.7 | 5.6 |
| Gemini 3.1 Pro Preview | 5.3 |
| GPT-5.4 Mini | 4.8 |
| Sonnet 4.6 | 4.0 |
| Kimi K2.6 | 4.0 |
| SWE-1.6 | 2.9 |
| MiniMax M2.7 | 2.7 |
| MiniMax M2.5 | 1.3 |
| Kimi K2.5 | 1.0 |
| Gemini 3.1 Flash-Lite Preview | 0.7 |