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

Add llama-cpp-python wheels with tensor cores support (#5003)

Browse source
This commit is contained in:
oobabooga 2023-12-19 17:30:53 -03:00 • committed by GitHub
parent 0a299d5959
commit de138b8ba6
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
9 changed files with 69 additions and 21 deletions
Download patch file Download diff file Expand all files Collapse all files

1
modules/shared.py
View file

@ -106,6 +106,7 @@ parser.add_argument('--compute_dtype', type=str, default='float16', help='comput
parser.add_argument('--quant_type', type=str, default='nf4', help='quant_type for 4-bit. Valid options: nf4, fp4.')
# llama.cpp
parser.add_argument('--tensorcores', action='store_true', help='Use llama-cpp-python compiled with tensor cores support. This increases performance on RTX cards. NVIDIA only.')
parser.add_argument('--n_ctx', type=int, default=2048, help='Size of the prompt context.')
parser.add_argument('--threads', type=int, default=0, help='Number of threads to use.')
parser.add_argument('--threads-batch', type=int, default=0, help='Number of threads to use for batches/prompt processing.')