Datasets:
Tasks:
Text Generation
Modalities:
Text
Formats:
parquet
Languages:
English
Size:
10K - 100K
ArXiv:
License:
Some SWE-rebench-V2 images keep newer releases of the project under test, and in some of them most of the reference patch's added code is readable
👀 1
2
#7 opened 3 days ago
by
rrrfffffeeee
348 tasks share a PR and base commit with a Multi-SWE-bench, SWE-PolyBench or SWE-bench Multilingual task (pair lists + filtered IDs)
👍 1
2
#5 opened about 1 month ago
by
raimondasl
Open-weights only toggle for Web UI
👍 1
#4 opened 8 months ago
by
TomLucidor
Sort by Pass@5 % on web UI
👍 1
#3 opened 8 months ago
by
TomLucidor
Can you add in Qwen3.5 and other series of models for testing?
3
#2 opened 8 months ago
by
TomLucidor
[bot] Conversion to Parquet
#1 opened 8 months ago
by
parquet-converter