[AMD] Fix QuantizedTensor import for ROCm TransformerEngine 2.8 - #58
Merged
guapisolo merged 2 commits intoJun 18, 2026
Conversation
…w-memory-resume patch QuantizedTensor moved from transformer_engine.pytorch.quantized_tensor to transformer_engine.pytorch.tensor in TE 2.x; try the new path and fall back to the old one so both TE layouts work. Co-authored-by: Zhiyao Jiang <jessicajiang324@gmail.com>
Co-authored-by: Zhiyao Jiang <jessicajiang324@gmail.com>
This was referenced Jul 24, 2026
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
ROCm TransformerEngine 2.8 moved
QuantizedTensorfromtransformer_engine.pytorch.quantized_tensortotransformer_engine.pytorch.tensor.This caused checkpoint loading with
--low-memory-resumeto fail withModuleNotFoundError.This change uses the new import path first and falls back to the old path for compatibility with other TransformerEngine versions. NVIDIA behavior is unchanged.
Testing
Verified on 8 MI355X GPUs with:
tests/e2e/ckpt/test_qwen3_4B_ckpt.pyThe complete save / load / async_save / load sequence passes. Both load stages report matching model hashes at iteration 2.
Before this change, the first load stage failed while importing
QuantizedTensor.