center4aai

A framework for measuring the robustness of LLM benchmarks — and of the models that run on them — under semantically-equivalent task reformulations.

Analytical CLI estimator for vector database disk and RAM sizing.

Local-first CLI for generating repository documentation from deterministic code facts with MkDocs publishing and optional OpenAI-compatible LLM support.

Unofficial Python client for the alphaxiv.org Assistant API — AI-powered Q&A over arXiv papers