Atlas

Benchmarks

← All benchmarks

FrontierCode 1.1 Extended - Pass Rate

Code · 2026-07-07

FrontierCode 1.1 Extended pass rate across the full set of 150 production-code tasks. A trial passes only when it satisfies every blocker criterion and is not flagged for unfair internet use.

Top models (higher is better)

ModelScore
Claude Fable 570.9
Claude Opus 569.6
GPT-5.6 Sol66.6
Opus 4.865.5
GPT-5.562.8
Grok 4.562.3
Sonnet 561.7
GPT-5.6 Terra61.7
GPT-5.6 Luna60.9
SWE-1.760.7
Opus 4.759.1
Kimi K2.7 Code50.0
Composer 2.545.1
GLM-5.244.1
DeepSeek-V4-Pro34.5
MiniMax M331.5
Inkling27.8
Qwen3.7-Plus24.1
SWE-1.622.7
Loading Atlas data…