oobabooga/text-generation-webui
1
0
Fork 0
mirror of https://github.com/oobabooga/text-generation-webui.git synced 2025-12-06 07:12:10 +01:00
Code Issues Actions Packages Projects Releases Wiki Activity

Remove some unnecessary code

Browse source
This commit is contained in:
oobabooga 2025-08-08 21:22:47 -07:00
parent 544c3a7c9f
commit 9e260332cc
1 changed files with 2 additions and 11 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

13
modules/exllamav3.py
View file

@ -26,6 +26,7 @@ from extensions.openai.image_utils import (
from modules import shared from modules import shared
from modules.logging_colors import logger from modules.logging_colors import logger
from modules.text_generation import get_max_prompt_length from modules.text_generation import get_max_prompt_length
from modules.torch_utils import clear_torch_cache
try: try:
import flash_attn import flash_attn
@ -342,6 +343,7 @@ class Exllamav3Model:
del self.model del self.model
except Exception as e: except Exception as e:
logger.warning(f"Error unloading main model: {e}") logger.warning(f"Error unloading main model: {e}")
self.model = None self.model = None
if hasattr(self, 'cache') and self.cache is not None: if hasattr(self, 'cache') and self.cache is not None:
@ -352,14 +354,3 @@ class Exllamav3Model:
if hasattr(self, 'tokenizer') and self.tokenizer is not None: if hasattr(self, 'tokenizer') and self.tokenizer is not None:
self.tokenizer = None self.tokenizer = None
# Force GPU memory cleanup
import gc
gc.collect()
if torch.cuda.is_available():
torch.cuda.empty_cache()
torch.cuda.synchronize()
torch.cuda.empty_cache()
logger.info("ExLlamaV3 model fully unloaded")