[HN Gopher] Gemlite: Towards Building Custom Low-Bit Fused CUDA ...
       ___________________________________________________________________
        
       Gemlite: Towards Building Custom Low-Bit Fused CUDA Kernels
        
       Author : un_ess
       Score  : 42 points
       Date   : 2024-08-15 14:23 UTC (8 hours ago)
        
 (HTM) web link (mobiusml.github.io)
 (TXT) w3m dump (mobiusml.github.io)
        
       | johnsutor wrote:
       | This would be great to have for the Triton language as well.
        
       | apsec112 wrote:
       | Weird that they don't mention Triton? I only skimmed it, but I'm
       | not sure what the pros and cons would be vs. Triton, which is the
       | tool I'd use if I wanted custom quantized inference kernels.
        
       ___________________________________________________________________
       (page generated 2024-08-15 23:01 UTC)