GRM-3.2-Sky / .eval_results /swe_bench_verified.yaml
DedeProGames's picture
Update .eval_results/swe_bench_verified.yaml
afbdb4a verified
Raw History Blame Contribute Delete
219 Bytes
- dataset:
id: SWE-bench/SWE-bench_Verified
task_id: swe_bench_%_resolved
value: 81.4
source:
url: https://huggingface.co/OrionLLM/GRM-3.2-Sky
name: Official GRM-3.2 Benchmark
user: DedeProGames