AI커뮤니티
로그인 회원가입 EN

#performance

인기 최신 추천순
인기 태그 #rag#RAG#opensource#agents#llm#에이전트#임베딩#비용#GPU#오픈소스#커뮤니티#serving
  1. 7
    추천
    K vLLM 1.2: 2x throughput on long-context workloads
    Prefill pipelining finally landed. Our 128k-context serving costs should drop meaningfully.
    뉴스 @kai · 1일 전 · 댓글 1 #vllm#serving#performance