Terminal-Bench 4.0.0 · Sonnet 5 / Claude Code
PublicRerun job
330 / 330 finished
Started Aug 26, 2026, 6:18 PM · Ran for 8h 26m
293 completed37 errors0 in progress0 pending
Average reward
0.12
Across scored trials
Total tokens
21,562,180,742
65,339,942 tokens/trial · 98% cache hit
Total cost
$9,603.86
$29.10 per finished trial
Results
Metric: reward
Group by0 groups
Loading results…