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

Agentic Coding on SWE-Bench Multilingual

71.7Accuracy

MiMo-V2-Flash

29.68440.59251.562.408Jan 6, 2026Jan 29, 2026Feb 21, 2026Mar 17, 2026Apr 9, 2026May 2, 2026May 26, 2026
Updated 6d ago

Evaluation Results

MethodLinks
2026.01
71.7
2026.01
70.2
2026.01
68
2026.05
67.2
2026.01
61.1
2026.05
60.3
2026.05
57.7
2026.05
55.7
2026.01
55.3
2026.05
51.7
2026.01
38.1
2026.01
37.2
2026.01
31.3