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

Add tensor split support for llama.cpp (#3171)

Browse source
This commit is contained in:
Shouyi 2023-07-26 07:59:26 +10:00 • committed by GitHub
parent f653546484
commit 031fe7225e
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
7 changed files with 20 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

7
modules/llamacpp_model.py
View file

@ -41,6 +41,12 @@ class LlamaCppModel:
cache_capacity = int(shared.args.cache_capacity)
logger.info("Cache capacity is " + str(cache_capacity) + " bytes")
if shared.args.tensor_split is None or shared.args.tensor_split.strip() == '':
tensor_split_list = None
else:
tensor_split_list = [float(x) for x in shared.args.tensor_split.strip().split(",")]
params = {
'model_path': str(path),
'n_ctx': shared.args.n_ctx,
@ -51,6 +57,7 @@ class LlamaCppModel:
'use_mlock': shared.args.mlock,
'low_vram': shared.args.low_vram,
'n_gpu_layers': shared.args.n_gpu_layers,
'tensor_split': tensor_split_list,
'rope_freq_base': 10000 * shared.args.alpha_value ** (64/63.),
'rope_freq_scale': 1.0 / shared.args.compress_pos_emb,
'n_gqa': shared.args.n_gqa or None,