ztxz16/fastllm
@ztxz16fastllm是后端无依赖的高性能大模型推理库。同时支持张量并行推理稠密模型和混合模式推理MOE模型，任意10G以上显卡即可推理满血DeepSeek。双路9004/9005服务器+单显卡部署DeepSeek满血满精度原版模型，单并发20tps；INT4量化模型单并发30tps，多并发可达60+。
Stars
4,954
Forks
485
Language
C++
License
Apache-2.0
Last push
1 week ago
Related intel (0)
No related intel yet
This repo has not appeared in any of the sources the radar tracks. The collector runs on a schedule — check back once it covers this repo.