inspect-evals

PyPI page
Author: UK AI Security Institute
Summary: Collection of large language model evaluations
Latest version: 0.25.0
Required dependencies: backoff | datasets | hf-xet | httpx | huggingface-hub | inspect-ai | jinja2 | numpy | pillow | pydantic | pyyaml | requests | tiktoken | toml
Optional dependencies: antlr4-python3-runtime | beautifulsoup4 | bioc | bs4 | cje-eval | click | deepdiff | docker | drain3 | faiss-cpu | filelock | gdown | gensim | google-genai | google-search-results | h5py | html2text | httpx | huggingface-hub | immutabledict | inspect-cyber | inspect-evals | inspect-swe | instruction-following-eval | jieba | jsonlines | langdetect | lxml | lxml-stubs | matplotlib | mpmath | nltk | odfpy | openai | openpyxl | overrides | pandas | pdf2image | pip | platformdirs | psutil | pydantic | pygithub | pypdf | python-dateutil | python-dotenv | python-pptx | pyyaml | rank-bm25 | rdchiral | rdkit | requests | rich | rouge | rouge-score | sacrebleu | scikit-learn | scipy | semgrep | sentence-transformers | swebench | sympy | tabulate | tenacity | tf-keras | tree-sitter | tree-sitter-cpp | types-beautifulsoup4 | types-docker | vimgolf

Downloads last day: 25,534
Downloads last week: 190,997
Downloads last month: 866,915
7-day smoothing