fix: address stable model export failures - #240
Merged
regisss merged 2 commits intoJul 21, 2026
Conversation
Merged
Summary: - Bound XLM-Roberta masked-LM dynamic sequence shapes to avoid ExecuTorch int_oo upper-bound failures. - Switch GPT-2 coverage to public openai-community/gpt2 and handle missing position_ids via cache_position. - Fix T5 dynamic encoder-length export by returning actual cross-attention K/V lengths instead of full static cross-cache buffers. Test plan: - GPT-2 focused slow test passes. - XLM-Roberta direct main_export(..., "xnnpack") passes. - T5 translation/summarization pass for both xnnpack and portable.
JakeStevens
force-pushed
the
fix-stable-model-export-failures
branch
from
July 20, 2026 20:00
34f6ce0 to
2fe5028
Compare
digantdesai
approved these changes
Jul 20, 2026
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
Collaborator
Author
|
Note: the failing tests are gemma3, llama, phi, phi4, qwen3. This PR resolved gp2, roberta, t5, which are all green now. I will address the rest in a follow up PR, still debugging. |
Collaborator
Sounds good. Can you just run |
Collaborator
Author
|
done, should pass lint now |
Collaborator
Author
|
Proposed fix for the remaining failures: |
regisss
approved these changes
Jul 21, 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
4 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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:
Bound XLM-Roberta masked-LM dynamic sequence shapes to avoid ExecuTorch int_oo upper-bound failures.
Switch GPT-2 coverage to public openai-community/gpt2 and handle missing position_ids via cache_position.
Fix T5 dynamic encoder-length export by returning actual cross-attention K/V lengths instead of full static cross-cache buffers.
Test plan:
GPT-2 focused slow test passes.
XLM-Roberta direct main_export(..., "xnnpack") passes.
T5 translation/summarization pass for both xnnpack and portable.