bump: torch/torchaudio from 2.8.0 to 2.10.0#1285
Open
Dinsmoor wants to merge 1 commit into
Open
Conversation
Bump PyTorch to 2.10.0 which includes bundled Triton 3.6.0 with proper aarch64 support. This enables native compilation (COMPILE=1) on NVIDIA GB10/Blackwell (sm_121a) hardware, reducing inference time from ~4-5 minutes to ~13 seconds on a 10-second clip.
6 tasks
Collaborator
|
This will cause an error produced by a depricated feature from torchaudio. |
Author
|
I wouldn't have noticed this being an issue as I only used the server mode to generate samples via api call
Feel free to take a stab at finding what replaced that (weird thing for torchAUDIO to stop doing.... Audio lol)
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Bump PyTorch to 2.10.0 which includes bundled Triton 3.6.0 with proper aarch64 support. This enables native compilation (COMPILE=1) on NVIDIA GB10/Blackwell (sm_121a) hardware, reducing inference time from ~4-5 minutes to ~13 seconds on a 10-second clip.
Is this PR adding new feature or fix a BUG?
Fix BUG.
Is this pull request related to any issue? If yes, please link the issue.
#1262
^ this guy, I had to do a hell of a workaround.
Other notes
A lot of work for a stupid dependency resolution.
Test with
COMPILE=1 docker compose --profile server build --no-cachethen run withCOMPILE=1 docker compose --profile server uplike normal, runs great on my dgx spark.