Atlas

Benchmarks

← All benchmarks

FrontierSWE: Dart → Haskell (Best@5)

Code · 2026-04-16

Reimplement the dart_style formatter as a standalone Haskell executable. FrontierSWE Best@5 is the highest percentage of verifier tests passed among up to five independent long-horizon agent runs.

Top models (higher is better)

ModelScore
Claude Fable 599.0
Opus 4.830.0
Grok 4.530.0
GPT-5.527.0
Opus 4.725.0
GLM-5.225.0
GPT-5.421.0
Gemini 3.1 Pro Preview11.0
GLM-5.18.9
Kimi K2.67.5
Composer 2.56.7
Opus 4.63.8
Kimi K2.53.8
DeepSeek-V4-Pro0.9
Qwen3.6 Plus (2026-04-02)0.8
Loading Atlas data…