Issue 2026-10-12 · Launches · 模型 · 开源airllmAirLLM 70B inference with a single 4GB GPU.GitHub Trending2 d agoRead original ↗Topics:airllm