char-fork/llama.cpp

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

John Balisfeat: cuda implementation for `ggml_conv_transpose_1d` (ggml/854)fde13b3bb

char/aliceai-t5-moe
158 B5 linesraw
1#include "common.cuh"
2
3#define CUDA_CONV_TRANPOSE_1D_BLOCK_SIZE 256
4
5void ggml_cuda_op_conv_transpose_1d(ggml_backend_cuda_context & ctx, ggml_tensor * dst);