2,294 real-world GitHub issues and their verified patches from 12 popular Python repositories. The gold standard benchmark for evaluating LLMs on practical software engineering tasks — writing actual code that resolves real bugs in production codebases.
| Provider | princeton-nlp |
| Category | 评估与基准测试 |
| Size | 2,294 Tasks |
| License | MIT |
| Downloads | 450k |
| Tags | Software Engineering, Real-world, GitHub, Benchmark, Code |
from datasets import load_dataset
ds = load_dataset("princeton-nlp/SWE-bench")