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

Add --load-in-4bit parameter (#2320)

Browse source
This commit is contained in:
oobabooga 2023-05-25 01:14:13 -03:00 • committed by GitHub
parent 63ce5f9c28
commit 361451ba60
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
6 changed files with 61 additions and 22 deletions
Download patch file Download diff file Expand all files Collapse all files

3
modules/ui.py
View file

@ -30,9 +30,10 @@ theme = gr.themes.Default(
def list_model_elements():
elements = ['cpu_memory', 'auto_devices', 'disk', 'cpu', 'bf16', 'load_in_8bit', 'wbits', 'groupsize', 'model_type', 'pre_layer', 'threads', 'n_batch', 'no_mmap', 'mlock', 'n_gpu_layers']
elements = ['cpu_memory', 'auto_devices', 'disk', 'cpu', 'bf16', 'load_in_8bit', 'load_in_4bit', 'compute_dtype', 'quant_type', 'use_double_quant', 'wbits', 'groupsize', 'model_type', 'pre_layer', 'threads', 'n_batch', 'no_mmap', 'mlock', 'n_gpu_layers']
for i in range(torch.cuda.device_count()):
elements.append(f'gpu_memory_{i}')
return elements