PyPI page
Home page
Author:
Michael Miller
Summary:
Rigorous quality benchmarking for local LLMs with paired statistics, LLM-as-judge scoring, and reproducible runs.
Latest version:
0.1.0
Required dependencies:
beaupy
|
httpx
|
ollama
|
pydantic
|
python-dotenv
|
pyyaml
|
rich
|
typer
Optional dependencies:
anthropic
|
pytest
|
pytest-asyncio
|
ruff
Downloads last day:
0
Downloads last week:
3
Downloads last month:
15
7-day smoothing