search

modelscope / dash-infer

DashInfer is a native LLM inference engine aiming to deliver industry-leading performance atop various hardware architectures, including x86 and ARMv9.
Apache License 2.0
130 stars 14 forks source link

汇编kernel是指x86和arm都有吗 #34

Closed LittleNoob2333 closed 1 week ago

LittleNoob2333 commented 2 months ago

zsbd

laiwenzh commented 1 month ago

x86、arm的部分kernel用汇编优化了，x86是一些小算子和flashattention，arm是小算子和gemm