mmarch/text-generation-webui
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

Add the --cpu option for llama.cpp to prevent CUDA from being used (#3432)

Browse source
This commit is contained in:
oobabooga 2023-08-03 11:00:36 -03:00 • committed by GitHub
parent 3e70bce576
commit 87dab03dc0
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
5 changed files with 40 additions and 10 deletions
Download patch file Download diff file Expand all files Collapse all files

17
modules/llamacpp_hf.py
View file

@ -10,13 +10,22 @@ from transformers.modeling_outputs import CausalLMOutputWithPast
from modules import shared
from modules.logging_colors import logger
import llama_cpp
if torch.cuda.is_available() and not torch.version.hip:
try:
from llama_cpp_cuda import Llama
import llama_cpp_cuda
except:
from llama_cpp import Llama
llama_cpp_cuda = None
else:
from llama_cpp import Llama
llama_cpp_cuda = None
def llama_cpp_lib():
if shared.args.cpu or llama_cpp_cuda is None:
return llama_cpp
else:
return llama_cpp_cuda
class LlamacppHF(PreTrainedModel):
@ -111,5 +120,7 @@ class LlamacppHF(PreTrainedModel):
'logits_all': True,
}
Llama = llama_cpp_lib().Llama
model = Llama(**params)
return LlamacppHF(model)