← Back to feed

Sturnus, a latency-based LLM load balancer

IT·开源 · 2026-10-26 15:20

Sturnus is an open-source LLM load balancer that distributes requests across multiple models or endpoints using latency-based scheduling. The available source material only captured GitHub navigation content, so implementation details remain unavailable.

Source: github.com


Auto-generated summary for reference only. RSS