FrontierCode 1.1 Main - Score
Code · 2026-07-07
FrontierCode 1.1 Main weighted score on the 100 hardest tasks in the 150-task Extended set. The 1.1 methodology adds fair-internet-use instructions and verification, and relaxes overly strict blocker criteria; flagged runs receive zero.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Opus 5.5 | 54.6 |
| Claude Fable 5 | 53.5 |
| GPT-6 Astra | 53.3 |
| Claude Sonnet 5.5 | 52.1 |
| Claude Fable 5.1 | 50.9 |
| GPT-6.1 Sol | 50.2 |
| SWE-2 | 50.0 |
| GPT-6 Sol | 49.3 |
| Grok 4.6 | 48.0 |
| Grok 4.7 | 47.6 |
| Gemini 4 Argon | 47.5 |
| GPT-5.6 Sol | 47.5 |
| Opus 4.8 | 46.5 |
| Claude Haiku 5.5 | 46.4 |
| DeepSeek V4.1 Flash | 43.9 |
| Gemini 3.7 Flash | 43.6 |
| GPT-5.5 | 43.0 |
| Sonnet 5 | 42.7 |
| Grok 4.5 | 42.4 |
| GPT-6 Luna | 42.4 |
| SWE-1.7 | 42.3 |
| GPT-5.6 Terra | 41.3 |
| Gemini 3.8 Flash | 41.2 |
| GLM-5.3 | 40.1 |
| GPT-5.6 Luna | 39.8 |
| Mistral Large 4 Preview | 39.4 |
| Opus 4.7 | 38.5 |
| Gemini 3.6 Flash | 34.4 |
| GLM-5.3 Flash | 31.8 |
| Kimi K2.7 Code | 30.1 |
| DeepSeek V4 Pro 0813 | 28.6 |
| GPT-5.4 Mini | 27.0 |
| Opus 4.6 | 26.6 |
| Composer 2.5 | 25.6 |
| GLM-5.2 | 24.5 |
| Sonnet 4.6 | 24.3 |
| DeepSeek-V4-Flash-0731 | 18.8 |
| DeepSeek-V4-Pro | 17.6 |
| MiniMax M3 | 14.7 |
| Inkling | 14.0 |