Original release notes
opencl: improve fa, allow dk512 for gemma-4, improve dk64 (#30266)
- opencl: enable Gemma-4 E4B GPU decode
Co-authored-by: Hongqiang Wang
- opencl: extend Gemma-4 GPU decode to E2B
Co-authored-by: Hongqiang Wang
- opencl: optimize DK64 GQA8 decode
Co-authored-by: Hongqiang Wang
- opencl: optimize DK128 GQA4 decode
Co-authored-by: Hongqiang Wang
---------
Co-authored-by: Hongqiang Wang
Website: -
Attestations: -
UI:

