Devotd Github

johnfink8/sfnx

johnfink8/sfnx 是一个把模型运算拆分到多块 PCIe 显卡上的推理引擎,其 README 公布了特定三块 V100 配置下的运行结果。

GitHub 项目说明: CUDA engine that splits every layer of a production quality MoE LLM across three PCIe V100 cards delivering 382 tok/s on a used $1,800 box.

在 GitHub 查看 ↗
星标
★ 4
复刻
0
语言
Python
许可证
MIT

文章

1