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

Software Engineering Issue Solving on SWE-Bench Verified

54Accuracy

OpenThinkerAgent-32B

4.70417.50230.343.098Dec 1, 2025Jan 4, 2026Feb 7, 2026Mar 13, 2026Apr 16, 2026May 20, 2026Jun 23, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
54-
2026.06
51-
2026.06
49.4-
2025.12
4627.39
2025.12
42.2-
2026.06
42.2-
2026.06
41.9-
2026.06
39.4-
2025.12
38.6-
2025.12
33.40
2025.12
32-
2026.06
29.1-
2025.12
8.815.79
2025.12
7.60
2025.12
6.6-