cuda : fix LLAMA_CUDA_F16 build

2024-03-25 15:31:10 +01:00
1 changed files with 1 additions and 0 deletions
--- a/ggml-cuda/dmmv.cu
+++ b/ggml-cuda/dmmv.cu
@ -1,5 +1,6 @@
 #include "dmmv.cuh"
 #include "dequantize.cuh"
+#include "convert.cuh"

 // dmmv = dequantize_mul_mat_vec
 #ifndef GGML_CUDA_DMMV_X