Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Software Engineering Tasks on SWE-bench Multilingual (test)
Loading...
77.8
Resolution Rate (%)
Claude Opus 4.6
69.896
71.948
74
76.052
Mar 29, 2026
Resolution Rate (%)
Updated 19d ago
Evaluation Results
Method
Method
Links
Resolution Rate (%)
Claude Opus 4.6
Scaffold=Claude Code
2026.03
77.8
KAT-Coder-V2
Scaffold=Claude Code
2026.03
75.4
KAT-Coder-V2
Scaffold=OpenCode
2026.03
71.2
Claude Opus 4.6
Scaffold=OpenCode
2026.03
70.2
Feedback
Search any
task
Search any
task