huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-04-21 06:42:10 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix(server): empty_cache when stopped

Browse Source
This commit is contained in:
OlivierDehaene 2023-07-15 13:57:31 +02:00
parent c58a0c185b
commit a2cf1bdb2f
1 changed files with 1 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
server/text_generation_server/models/flash_causal_lm.py
View File

@ -991,6 +991,7 @@ class FlashCausalLM(Model):
if stopped: if stopped:
del batch del batch
torch.cuda.empty_cache()
# No need to return a batch if we know that all requests stopped # No need to return a batch if we know that all requests stopped
return generations, None return generations, None