huggingface/candle
1
0
Fork 0
mirror of https://github.com/huggingface/candle.git synced 2025-06-19 19:58:35 +00:00
Code Issues Packages Projects Releases Wiki Activity

Add the bf16 cuda kernels.

Browse Source
This commit is contained in:
laurent
2023-06-29 23:12:02 +01:00
parent 018e017e7e
commit ec79fc43f2
9 changed files with 67 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
candle-kernels/src/reduce.cu
View File

@ -43,6 +43,10 @@ extern "C" __global__ void FN_NAME( \
} \
} \
#if __CUDA_ARCH__ >= 800
SUM_OP(__nv_bfloat16, sum_bf16)
#endif
#if __CUDA_ARCH__ >= 530
SUM_OP(__half, sum_f16)
#endif