Atlas

Benchmarks

← All benchmarks

SWE Atlas - Test Writing

Code · 2026-05-07

Score on SWE Atlas test-writing tasks that write production-grade tests for repository behavior.

Top models (higher is better)

ModelScore
Claude Fable 555.6
Opus 4.849.6
GPT-5.6 Sol45.9
GPT-5.444.4
GPT-5.542.6
GLM-5.241.5
Muse Spark 1.141.5
GPT-5.3-Codex39.0
Opus 4.738.5
Opus 4.636.7
Sonnet 4.631.8
Muse Spark31.1
Gemini 3 Flash Preview30.3
Gemini 3.1 Pro Preview29.8
GLM-528.7
DeepSeek-V4-Pro27.1
Kimi K2.525.8
MiniMax M2.518.6
Loading Atlas data…