Share your thoughts, 1 month free Claude Pro on usSee more

Agentic Coding on R&D Coding Benchmark

80Pass Rate

Opus 4.6

Updated 1mo ago

Evaluation Results

Method	Links
Opus 4.6 2026.04		80
Opus 4.5 2026.04		73
Opus 4.5 2026.04		70
DeepSeek-V4-Pro-Max 2026.04		67
Sonnet 4.5 2026.04		47
Haiku 4.5 2026.04		13