📚 Wilson's Knowledge Base

标签: latency

此标签下有1条笔记。

  • 2026年9月12日

    How continuous batching enables 23x throughput in LLM inference while reducing p50 latency

    • continuous-batching
    • llm-inference
    • vllm
    • paged-attention
    • throughput
    • latency
    • benchmarking
    • orca

Created with Quartz v4.5.2 © 2026 · 部署于 2026/10/08 07:59

  • GitHub
  • RSS