Mirrors/text-generation-webui
1
0
Fork 0
mirror of https://github.com/oobabooga/text-generation-webui.git synced 2024-11-22 08:07:56 +01:00
Code Issues Packages Projects Releases Wiki Activity

add n_batch support for llama.cpp (#1115)

Browse Source
This commit is contained in:
eiery 2023-04-24 02:46:18 -04:00 committed by GitHub
parent 2f6e2ddeac
commit 78d1977ebf
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
3 changed files with 4 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
README.md
View File

@ -220,6 +220,7 @@ Optionally, you can use the following command-line flags:
| Flag | Description |
|-------------|-------------|
| `--threads` | Number of threads to use in llama.cpp. |
| `--n_batch` | Processing batch size for llama.cpp. |
#### GPTQ

3
modules/llamacpp_model_alternative.py
View File

@ -24,7 +24,8 @@ class LlamaCppModel:
'model_path': str(path),
'n_ctx': 2048,
'seed': 0,
'n_threads': shared.args.threads or None
'n_threads': shared.args.threads or None,
'n_batch': shared.args.n_batch
}
self.model = Llama(**params)
self.model.set_cache(LlamaCache)

1
modules/shared.py
View File

@ -119,6 +119,7 @@ parser.add_argument('--trust-remote-code', action='store_true', help="Set trust_
# llama.cpp
parser.add_argument('--threads', type=int, default=0, help='Number of threads to use in llama.cpp.')
parser.add_argument('--n_batch', type=int, default=8, help='Processing batch size for llama.cpp.')
# GPTQ
parser.add_argument('--wbits', type=int, default=0, help='Load a pre-quantized model with specified precision in bits. 2, 3, 4 and 8 are supported.')