v4.0: Production-grade optimizations - priority queue, prefix caching, TTL, metrics, TTFT tracking c9737d6 Matrix Agent commited on Dec 10, 2025
v3.2: Speed optimizations - OpenBLAS, dual models (7B/1.5B), model selector, timing display 1b50d66 Matrix Agent commited on Dec 10, 2025
Add frontend dashboard, comprehensive docs, and enhanced logging v3.1 8910367 Matrix Agent commited on Dec 10, 2025
Fix asyncio event loop error in QueuedRequest dataclass c5b1f8b Matrix Agent commited on Dec 10, 2025