swebench
The official SWE-bench package - a benchmark for evaluating LMs on software engineering
- Category
- Python packages
pip install swebench Official docs outline
github.comExtra Credit builds an outline of swebench's docs the first time your work uses it. Until then, start at the official documentation.
More in Python packages
All 9,553-
boltons When they're not builtins, they're boltons.
-
pytimeparse Time expression parser
-
agent-client-protocol A Python implement of Agent Client Protocol (ACP, by Zed Industries)
-
lupa Python wrapper around Lua and LuaJIT
-
eval-type-backport Like `typing._eval_type`, but lets older Python versions use newer typing features.
-
logfire-api Shim for the Logfire SDK which does nothing unless Logfire is installed
-
grimp Builds a queryable graph of the imports within one or more Python packages.
-
opentelemetry-instrumentation-aiohttp-client OpenTelemetry aiohttp client instrumentation
-
pytest-metadata pytest plugin for test session metadata
-
retrying Retrying
-
vcrpy Automatically mock your HTTP interactions to simplify and speed up testing
-
parsedatetime Parse human-readable date/time text.