- dataset:
    id: ScaleAI/SWE-bench_Pro
    task_id: SWE_Bench_Pro
  value: 54.0
  date: '2026-04-23'
  source:
    url: https://huggingface.co/OrionLLM/GRM-2.6-Plus
    name: Official GRM-2.6 Benchmark
    user: DedeProGames