We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
There was an error while loading. Please reload this page.
2 parents 3e7eae4 + 282698b commit 0952d53Copy full SHA for 0952d53
llama_cpp/server/app.py
@@ -30,6 +30,9 @@ class Settings(BaseSettings):
30
ge=0,
31
description="The number of layers to put on the GPU. The rest will be on the CPU.",
32
)
33
+ seed: int = Field(
34
+ default=1337, description="Random seed. -1 for random."
35
+ )
36
n_batch: int = Field(
37
default=512, ge=1, description="The batch size to use per eval."
38
@@ -109,6 +112,7 @@ def create_app(settings: Optional[Settings] = None):
109
112
llama = llama_cpp.Llama(
110
113
model_path=settings.model,
111
114
n_gpu_layers=settings.n_gpu_layers,
115
+ seed=settings.seed,
116
f16_kv=settings.f16_kv,
117
use_mlock=settings.use_mlock,
118
use_mmap=settings.use_mmap,
0 commit comments