Package profile
swebench
- Summary: The official SWE-bench package - a benchmark for evaluating LMs on software engineering
- Author: SWE-bench team <support@swebench.com>
- License: MIT License
- Source: https://github.com/swe-bench/SWE-bench
- Number of releases: 119
- First release: 0.1 on 2023-11-01
- Latest release: 5.0.2 on 2026-08-18
- Latest release size: 139.4 KB (pure Python wheel)
Appears in:Packages with highest PyPI growth in April 2026Packages with highest PyPI growth in August 2026Packages with highest PyPI growth in March 2026
Dependencies
Swebench has 37 dependencies, 21 of which optional.Dependent packages
| Package | Optional | Group |
|---|---|---|
| inspect-evals | true | swe-bench |
| swesmith | true | all |
Similar packages
- pydantic-evalsFramework for evaluating stochastic code execution, especially code making use of LLMs
- swesmithThe official SWE-smith package - A toolkit for generating software engineering training data at scale.
- deepevalThe LLM Evaluation Framework
- langgraphBuilding stateful, multi-actor applications with LLMs
- vllmA high-throughput and memory-efficient inference and serving engine for LLMs
- cloudml-hypertuneA library to report Google CloudML Engine HyperTune metrics.
- swigSWIG is a software development tool that connects programs written in C and C++ with a variety of high-level programming languages.