Skip to content

Fix GPU memory leak in voice conditioning cache #166

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
secretzer0 wants to merge 1 commit into from
Closed

Fix GPU memory leak in voice conditioning cache #166

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion engine.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -487,7 +487,7 @@ def synthesize(
# Call the core model's generate method.
# autocast promotes float32 inputs to bfloat16 to match T3/S3Gen weights,
# keeping numerically sensitive ops (softmax, norms) in float32 automatically.
with torch.autocast("cuda", dtype=torch.bfloat16, enabled=BF16_ENABLED):
with torch.inference_mode(), torch.autocast("cuda", dtype=torch.bfloat16, enabled=BF16_ENABLED):
if loaded_model_type == "multilingual":
wav_tensor = chatterbox_model.generate(
text=text,
Expand Down