大模型推理案头手册（开源版）：给定 GPU 算力，一个模型能扛多少 QPS？三层模型 × 三面墙 × 排队论 × 开环压测
shtjww/llm-inference-capacity-handbook: 大模型推理案头手册（开源版）：给定 GPU 算力，一个模型能扛多少 QPS？三层模型 × 三面墙 × 排队论 × 开环压测
Related
Alan-Heoooh/MATE: MATE: Asynchronous Multi-Rate Diffusion Policy Composition
MATE: Asynchronous Multi-Rate Diffusion Policy Composition
duwqijlk/muse-protocol: muse.ai 注册与邀请码兑换协议研究（纯 HTTP Python 实现）
muse.ai 注册与邀请码兑换协议研究（纯 HTTP Python 实现）
briskember13/adobe-acrobat-pdf-desktop: Adobe Acrobat PDF 2026 desktop setup with text and image editing directly inside any PDF, conversion to and from Word, Excel and images, plus fill, sign and merge documents with an AI assistant for Windows 11
Adobe Acrobat PDF 2026 desktop setup with text and image editing directly inside any PDF, conversion to and from Word, Excel and images, plus fill, sign and merge documents with an...