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

Software Engineering on SWE-bench Verified (Success Rate (%))

80.9Success Rate

Claude-Opus-4.5

-3.23618.60740.4562.293Apr 13, 2026Apr 23, 2026May 4, 2026May 15, 2026May 25, 2026Jun 5, 2026Jun 16, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
80.9
2026.06
76.3
76.2
2026.06
73.8
73.1
2026.04
71.8
2026.04
70.4
2026.04
69.6
69.2
67
2026.04
66.6
2026.04
65.4
2026.06
64.4
2026.04
64.2
60.4
2026.04
50.4
2026.06
50
2026.04
49.3
2026.04
48.6
45.2
2026.05
43.7
2026.06
43
2026.05
42.3
2026.04
42.2
2026.04
41
2026.04
40.2
2026.04
38.8
2026.04
38
2026.04
36.6
2026.04
34.4
2026.04
32.8
2026.04
30.06
2026.06
27.6
2026.04
26.8
2026.06
24.5
2026.04
24.1
2026.04
23.4
2026.06
22.4
2026.04
21.6
2026.04
21.45
2026.04
20.6
2026.06
20
2026.06
20
2026.04
19
2026.04
16.4
2026.04
15.83
2026.04
15.2
2026.04
10.6
0
0
0
0
0