FrontierCode 1.1 Main - Pass Rate
Code · 2026-07-07
FrontierCode 1.1 Main pass rate on the 100 hardest tasks in the 150-task Extended set. A trial passes only when it satisfies every blocker criterion and is not flagged for unfair internet use. The result averages repeated trials per task.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Opus 5.5 | 60.0 |
| Claude Fable 5 | 58.9 |
| Claude Opus 5 | 58.9 |
| GPT-6 Astra | 58.8 |
| Claude Sonnet 5.5 | 57.2 |
| GPT-6.1 Sol | 55.8 |
| Claude Fable 5.1 | 55.6 |
| SWE-2 | 55.5 |
| GPT-6 Sol | 54.3 |
| Grok 4.6 | 53.1 |
| Grok 4.7 | 53.1 |
| GPT-5.6 Sol | 52.9 |
| Gemini 4 Argon | 52.8 |
| Claude Haiku 5.5 | 51.6 |
| Opus 4.8 | 51.6 |
| Gemini 3.7 Flash | 48.9 |
| DeepSeek V4.1 Flash | 48.6 |
| GPT-5.5 | 48.2 |
| GPT-6 Luna | 47.8 |
| SWE-1.7 | 47.8 |
| Sonnet 5 | 47.6 |
| Grok 4.5 | 47.2 |
| Gemini 3.8 Flash | 46.7 |
| GPT-5.6 Terra | 46.3 |
| GPT-5.6 Luna | 44.7 |
| GLM-5.3 | 44.7 |
| Mistral Large 4 Preview | 43.5 |
| Opus 4.7 | 42.8 |
| Gemini 3.6 Flash | 38.9 |
| GLM-5.3 Flash | 35.7 |
| Kimi K2.7 Code | 33.6 |
| DeepSeek V4 Pro 0813 | 31.8 |
| GPT-5.4 Mini | 30.8 |
| Opus 4.6 | 29.7 |
| Composer 2.5 | 29.3 |
| Sonnet 4.6 | 27.5 |
| GLM-5.2 | 27.4 |
| DeepSeek-V4-Flash-0731 | 21.1 |
| DeepSeek-V4-Pro | 20.0 |
| MiniMax M3 | 16.6 |