PyPI page
Home page
Author:
None
License:
MIT
Summary:
A multi-dimensional behavioral framework for evaluating LLM reasoning quality beyond accuracy
Latest version:
1.1.4
Required dependencies:
anthropic
|
bert-score
|
datasets
|
google-generativeai
|
numpy
|
openai
|
openpyxl
|
python-dotenv
|
pyyaml
|
requests
|
sentence-transformers
|
tqdm
Optional dependencies:
accelerate
|
bitsandbytes
|
llm-reasoning-quality
|
matplotlib
|
seaborn
|
sentencepiece
|
torch
|
transformers
Downloads last day:
11
Downloads last week:
74
Downloads last month:
659