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

Terminal-based agent interaction on Terminal-Bench 2.0

59.3Success Rate (SR)

Claude-Opus-4.5

-2.37213.63929.6545.661Jun 11, 2026Jun 12, 2026Jun 13, 2026Jun 14, 2026Jun 15, 2026Jun 16, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
59.3-
54.2-
2026.06
47.6-
46.4-
2026.06
41-
2026.06
33.71.23
2026.06
31.52.2
2026.06
30.32.31
2026.06
29.26.32
27.8-
2026.06
279.59
2026.06
272.61
2026.06
24.74.19
23.6-
2026.06
21-
2026.06
20.20.42
2026.06
19.11.87
2026.06
14.61.33
2026.06
13.51.28
13.5-
2026.06
12.2-
2026.06
11.21.6
2026.06
11.2-
2026.06
10.11.41
2026.06
9-
2026.06
6.74.9
4.5-
2.5-
2.3-
0-
0-
0-