Search

TUI for browsing, searching, and resuming Claude Code sessions

The Obsidian that populates itself so your Claude never forgets.

High-performance, memory-fluid LLM inference engine — Rust speed, Python convenience.

Smart AI router — filters requests to the right model, saving 40-80% on costs

Efficient HTTP blob storage over local filesystem

A comprehensive, production-ready LLM security and guardrails framework

Adaptive Streaming Vector Quantization KV Cache for LLMs