char-fork/llama.cpp

git clone https://git.t4t.associates/char-fork/llama.cpp

Molly Sophiallama: add support for QRWKV6 model architecture (#11001)ee7136c6d2025-01-10

#include "common.cuh"

void ggml_cuda_op_gated_linear_attn(ggml_backend_cuda_context & ctx, ggml_tensor * dst);