Last released Apr 25, 2026
VRAM-aware LLM routing daemon — local-first, OpenAI-compatible
Supported by