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

Software Engineering Agent Task on SWE-Bench Pro (pass@3)

100Pass@3

LLM Agent (forced-injection)

43.8458.427387.58May 8, 2026
Updated 23d ago

Evaluation Results

MethodLinks
2026.05
100
2026.05
92
2026.05
92
2026.05
92
2026.05
92
2026.05
92
2026.05
86
2026.05
83
2026.05
83
2026.05
81
2026.05
81
2026.05
81
2026.05
81
2026.05
79
2026.05
79
2026.05
75
2026.05
75
2026.05
74
2026.05
74
2026.05
74
2026.05
71
2026.05
68
2026.05
67
2026.05
67
2026.05
67
2026.05
63
2026.05
56
2026.05
46