llama.cpp icon indicating copy to clipboard operation
llama.cpp copied to clipboard

Published 20 hours ago • verified publisher icon ggml-org

opencl: fix for small models

Open lhez opened this issue 4 days ago • 0 comments

Currently small models like qwen2.5 0.5B does not work properly with OpenCL backend. This PR fixes this issue. This PR also changes subgroup size to 64 for all Adreno GPUs.

lhez avatar Feb 18 '25 23:02 lhez