.eval_results/swe_bench_verified.yaml

- dataset:
    id: SWE-bench/SWE-bench_Verified
    task_id: swe_bench_%_resolved
  value:  74.8
  source:
    url: https://huggingface.co/papers/2603.16790
    name: IndustrialCoder technical report
    user: nielsr