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

Agentic on Terminal Bench hard

25.5Accuracy

Super

22.79623.49824.224.902Jul 5, 2026
Updated 18d ago

Evaluation Results

MethodLinks
2026.07
25.5
2026.07
24.9
2026.07
24
2026.07
23.4
2026.07
22.9