llama.cpp

et-backend: Q8_0 matrix-engine GEMM kernel for prefill

#25661

Open