huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-09-11 04:14:52 +00:00
Code Issues Packages Projects Releases Wiki Activity

Let each model resolve their own default dtype.

Browse Source
This commit is contained in:
Nicolas Patry 2023-11-27 10:30:35 +00:00
parent ed2a3f617e
commit 2713b21132
1 changed files with 3 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
server/text_generation_server/models/__init__.py
View File

@ -87,7 +87,9 @@ def get_model(
trust_remote_code: bool,
) -> Model:
if dtype is None:
dtype = torch.float16
# Keep it as default for now and let
# every model resolve their own default dtype.
dtype = None
elif dtype == "float16":
dtype = torch.float16
elif dtype == "bfloat16":