Inkling | SWE-bench Verified — bash-only harness
Thinking Machines Lab's Inkling evaluation results on SWE-bench Verified — bash-only harness
Score: 77.6%
Reasoning level: effort=0.99
Thinking Machines Lab's Inkling evaluation results on SWE-bench Verified — bash-only harness
Score: 77.6%
Reasoning level: effort=0.99