llama.cpp
51571722 - opencl: optimize mul_mat_f16_f32_l4 for decode (#24504)

Commit

42 days ago

opencl: optimize mul_mat_f16_f32_l4 for decode (#24504)

References

#24504 - opencl: optimize mul_mat_f16_f32 for decode

Author

lhez

lhez

Parents

Loading