forked from AI-Hypercomputer/maxdiffusion
-
Notifications
You must be signed in to change notification settings - Fork 1
Enable flux multinode #1
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.
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
cpersson-amd
wants to merge
19
commits into
jianhan/synthetic_data_iterator
from
cpersson/flux_multinode
Closed
Enable flux multinode #1
cpersson-amd
wants to merge
19
commits into
jianhan/synthetic_data_iterator
from
cpersson/flux_multinode
Conversation
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
…-Hypercomputer#311) - Rename VaceWanPipeline to VaceWanPipeline2_1. - Make VaceWanPipeline2_1 inherit from WanPipeline2_1. - Remove calls to self.transformer2. This change aligns the class with the WanPipeline naming structure and resolves an initialization bug when using the from_pretrained method.
…puter#312) - ubuntu-20.04 has been deprecated - using ubuntu-latest for robustness
* Fix formatting with pyink * Update Python and pyink version to be aligned with code_style.sh
…puter#299) * add flash attn te support for wan * add gpu optimized sharding parallelism * sharding bugfixes * generalize across sharding parallelisms * fix issue with inference using fsdp + te flash attention * revert fsdp_tpu name change * update readme with wan2.1 gpu notes * re-order parallelism axes and revert dynamic context parallel axes selection * remove unused max_utils imports * change mesh names to more accurately reflect sharding * cleanup * fix lint errors * update configs for unit tests. --------- Co-authored-by: Juan Acevedo <juancevedo@gmail.com>
…#313) * sharding before call to vae encoder * sharding before call to vae encoder * ruff check * pyink checks * pyink check
In AI-Hypercomputer#306 the return of WanTimeTextImageEmbedding was modified to include a 5th element. This refactors the call in WanVACEModel to comply with the new interface
* fp8 bug * format * unit test * unit test
…#316) * Adding check for batch size divisibility before sharding video condition tensor * pyink checks * Removed unused var * Moving commit retrieval to before JAX setup init * fix * replaced boundary_timestep with ratio in wan2.2 t2v * replaced boundary_timestep with ratio in wan2.2 t2v
AI-Hypercomputer#322) * Move the top level 'tests/' into `src/maxdiffusion/tests/` as a legacy * Format by pyink
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
No description provided.