Last released May 3, 2026
Distributed LLM inference on Apple Silicon — Pipeline Parallelism + Speculative Decoding + OpenAI-compatible API
Multi-Agent governance framework — voting, veto, precedent, and risk classification
Supported by