#numa (3 Repositories)
Ranked open-source repositories tagged with #numa, scored by pull request acceptance likelihood and maintainer engagement velocity.
56.8%
39.4h
3 repositories tagged #numa
guqiong96/Lsglang
Lsglang is a special extension of sglang that fully utilizes CPU and GPU computing resources with an efficient GPU parallel + NUMA parallel architecture, suitable for MOE model hybrid inference.
Scottcjn/ram-coffers
LLM infrastructure cost reduction via NUMA-aware weight banking: 147 t/s (8.8x stock llama.cpp) on refurbished enterprise POWER8. Self-hosted inference, no cloud APIs. Part of the Proof of Physical AI stack.
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.