xinyun/ComfyUI-MochiWrapper
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Update mz_gguf_loader.py

Browse Source
This commit is contained in:
kijai 2024-10-28 04:07:11 +02:00
parent db23e2ecc0
commit ce903c0384
1 changed files with 1 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
mz_gguf_loader.py
View File

@ -27,6 +27,7 @@ def quantize_load_state_dict(model, state_dict, device="cpu", cublas_ops=False):
setattr(model, "cublas_half_matmul", True) setattr(model, "cublas_half_matmul", True)
print("Using cublas_ops") print("Using cublas_ops")
except: except:
print("Failed to load cublas_ops")
raise ImportError("Install cublas_ops (https://github.com/aredden/torch-cublas-hgemm) to use cublas_ops") raise ImportError("Install cublas_ops (https://github.com/aredden/torch-cublas-hgemm) to use cublas_ops")
else: else:
linear_ops = F.linear linear_ops = F.linear