"git@developer.sourcefind.cn:orangecat/ollama.git" did not exist on "f91bb2f7f00e5c82b30f08f56fd99b6f7c735e00"
  • zjing14's avatar
    Batched GEMM for fp16 (#79) · b53e9d08
    zjing14 authored
    * prepare host for batched_gemm
    
    * init commit of batched kernels
    
    * fixed
    
    * refine transform with freeze
    
    * m/n padding
    
    * fixed a bug; clean
    
    * add small tiles
    
    * clean
    
    * clean code
    
    * clean code
    
    * add nt, tn, tt layout
    
    * add missing file
    
    * use StaticBufferTupleOfVector instead
    
    * add reference_batched_gemm
    
    * fixed a macro
    b53e9d08
reference_batched_gemm.hpp 3.9 KB