.eval_results/swe_bench_verified.yaml

- dataset:
    id: SWE-bench/SWE-bench_Verified
    task_id: swe_bench_%_resolved
  value: 77.7
  date: '2026-04-23'
  source:
    url: https://huggingface.co/OrionLLM/GRM-2.6-Plus
    name: Official GRM-2.6 Benchmark
    user: DedeProGames