vrr/kvcache-ai-ktransformers
2
0
Fork 0
mirror of https://github.com/kvcache-ai/ktransformers.git synced 2025-09-13 00:29:59 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 2

fix KExpertsMarlin on GPU with out CUDA Graph

Browse source
This commit is contained in:
Atream 2025-02-24 09:30:54 +00:00
parent f5f6c6b95d
commit f327695079
2 changed files with 13 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

2
ktransformers/util/custom_gguf.py
View file

@ -310,6 +310,8 @@ class GGUFLoader:
values = GGML_DEQUANTIZE[ggml_name](data)
values = torch.from_numpy(values.copy())
if ggml_name == "BF16":
values = values.view(torch.bfloat16)
values = values.view(shape[-2::-1])
return values