FrontierCode 1.1 Extended - Score
Code · 2026-07-07
FrontierCode 1.1 Extended weighted score across the full set of 150 production-code tasks. 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 Fable 5 | 64.9 |
| Claude Opus 5 | 63.6 |
| GPT-5.6 Sol | 60.6 |
| Opus 4.8 | 59.6 |
| GPT-5.5 | 56.7 |
| Grok 4.5 | 56.6 |
| Sonnet 5 | 56.2 |
| GPT-5.6 Terra | 55.8 |
| GPT-5.6 Luna | 55.1 |
| SWE-1.7 | 54.6 |
| Opus 4.7 | 53.9 |
| Kimi K2.7 Code | 45.4 |
| Composer 2.5 | 40.8 |
| GLM-5.2 | 40.0 |
| DeepSeek-V4-Pro | 31.0 |
| MiniMax M3 | 28.5 |
| Inkling | 24.8 |
| Qwen3.7-Plus | 21.8 |
| SWE-1.6 | 20.5 |