PyPI page
Home page
Author:
Stefan Song
Summary:
Friendly evaluation and regression-testing framework for AI agents: inspectable traces, graded outcomes, baseline comparisons, and CI-ready reliability signals.
Latest version:
0.7.0
Required dependencies:
jsonschema
|
numpy
|
pydantic
|
pyyaml
|
scipy
Optional dependencies:
anthropic
|
datasets
|
httpx
|
mkdocs-material
|
mkdocstrings
|
mypy
|
openai
|
pytest
|
pytest-asyncio
|
pytest-cov
|
ruff
|
scipy-stubs
|
types-jsonschema
|
types-pyyaml
Downloads last day:
8
Downloads last week:
105
Downloads last month:
428
7-day smoothing