model

Retrieval and Retrieval-augmented LLMs

A vector index built on TurboQuant, written in Rust with Python bindings

Adding guardrails to large language models.

Fast and Accurate Code Search for Agents. Uses ~98% fewer tokens than grep+read