PrototypeCommon
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
- Owner
- shtjww (personal account)
- Stars
- Stars: 185
- Forks
- Forks: 0
- Language
- No language
- License
- None
- Created
- 2026
- Topics
- capacity-planninggpullm-inferencellmopsqueueing-theoryvllm
- Tier
- Prototype (100 to 249 ★) What this tier means