Scale AI
@scale-ai
⌘K
scale-ai/swe-atlas-qna
latestrev. 1
SWE-Atlas - Codebase QnA is a benchmark of deep codebase comprehension and QnA problems for coding agents. Checkout https://github.com/scaleapi/SWE-Atlas/ for instructions on running it.
124 tasksMohit Raghavendra
scale-ai/swe-atlas-tw
latestrev. 2
SWE-Atlas - Test Writing -- A benchmark of comprehensive test writing problems for coding agents. Checkout https://github.com/scaleapi/SWE-Atlas/ for instructions on running it.
90 tasksMohit Raghavendra
scale-ai/swe-bench-pro
latestrev. 2
731 tasks
3 datasets