AlphaUncommon
guqiong96/Lvllm
LvLLM is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models.
- Owner
- guqiong96 (personal account)
- Stars
- Stars: 474
- Forks
- Forks: 42
- Language
- Python
- License
- Apache-2.0
- Created
- 2025
- Topics
- hybridinferencemodelmoenumaparallelismvllmcpudecodegpu
- Tier
- Alpha (250 to 699 ★) What this tier means
More repos by guqiong96
All 2 repos by guqiong96- guqiong96/LsglangStars: 164PythonPrototype
Nearby in Python
All Python repos- QuZhan51496/paper2anythingStars: 474PythonAlpha
- ByteDance-Seed/EdgeBenchStars: 474PythonAlpha
- chengmarc/minecraft-citygenStars: 474PythonAlpha
- django-commons/django-robotsStars: 474PythonAlpha
- py-moneyed/py-moneyedStars: 474PythonAlpha
- TabViewer/tabviewStars: 474PythonAlpha