Skip to the page
/
Senior Engineer Benchmark
Proof rewrite, integrated
Scoreboard
Models
8
Claude Opus 5.5
Claude Sonnet 5.5
GPT-6 Astra
GPT-6.1 Sol
Muse Wonder Full 1.0
Muse Wonder 1.0
GPT-6 Sol
Gemini 4 Argon
Muse Wonder Full 1.0
47%
overall
By task
Model
Overall
Proof rewrite, integrated
Muse Wonder Full 1.0
Muse Code · high effort
47%
47%
×2 44–50%