2 packages found
Self-hosted, local-first AI assistant server: swappable LLM providers, review-gated memory, an encrypted vault, and wake
Compression with a quality contract — cache-aware, causally-pruned LLM context compression for agentic runtimes, certifi