Fix WAN training/inference error for new JAX/Flax Version#265
Merged
Conversation
entrpn
approved these changes
Oct 9, 2025
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The newer version of JAX 0.7.2 and Flax 0.12.0 now strictly requires a mesh to be defined whenever you initialize parameters with sharding rules, even in a single-device unit test environment.
From Flax team, the issue is due to this change: https://github.com/google/flax/blob/main/docs_nnx/flip/4844-var-eager-sharding.md
Simplify the creation of sharded NNX models. When a sharding annotation is provided, all nnx.Variable creation will require a mesh context and automatically be sharded as annotated.
It can be disabled by using flax.config.update('flax_always_shard_variable', False).
In maxDiffusion, only WAN 2.1 use the flax modules.