Arm backend: Add serialized xlarge TOSA model suite - #21492
Merged
bdemirb merged 2 commits intoJul 30, 2026
Conversation
The DeepSeek-R1-Distill-Qwen layer-test PRs were reverted because they caused OOMs in the no-driver Arm TOSA model pytest job. The normal TOSA model shard should keep using pytest-xdist auto parallelism so unrelated model tests do not slow down. Add a separate serialized suite for xlarge TOSA model tests instead. The reland of the DeepSeek layer tests can mark the memory-heavy cases as xlarge and route them through this suite. This also gives internal CI an explicit option for running any xlarge TOSA model coverage serially. Change-Id: I83c09ce0ed73f131e1d02d4c6e526c3ce53188c0 Signed-off-by: Baris Demir <baris.demir@arm.com>
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/21492
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 873b1b9 with merge base b26b9ac ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
meta-cla
Bot
added
the
CLA Signed
Collaborator
Author
|
@pytorchbot label "partner: arm" |
pytorch-bot
Bot
added
the
partner: arm
Collaborator
Author
|
@pytorchbot label "release notes: arm" |
Sebastian-Larsson
approved these changes
Jul 30, 2026
bdemirb
added a commit
to bdemirb/executorch
that referenced
this pull request
Jul 30, 2026
Relands the DeepSeek-R1-Distill-Qwen-1.5B layer tests that were reverted by pytorch#21048 because the TOSA model shard could run several checkpoint-shaped exports concurrently and OOM. The prerequisite CI change in pytorch#21492 added a serialized xlarge TOSA model suite while keeping the normal TOSA model shard parallelized. This reland marks the DeepSeek TOSA layer tests as xlarge so they are excluded from the normal shard and can be routed through the serialized xlarge suite. The tests use the checkpoint configuration from the Hugging Face model and the upstream Qwen2 layer implementations that back this distilled model. The covered layers include rotary embedding, rotary application, KV repetition, attention, RMSNorm, MLP, decoder layer, and final norm. Token embedding is excluded because the full checkpoint embedding allocation is too large for regular CI. Signed-off-by: Baris Demir <baris.demir@arm.com> Change-Id: Ia28581bbb4ffe070bc35af060fcceef2ac90084a
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 DeepSeek-R1-Distill-Qwen layer-test PRs were reverted because they caused OOMs in the no-driver Arm TOSA model pytest job.
The normal TOSA model shard should keep using pytest-xdist auto parallelism so unrelated model tests do not slow down. Add a separate serialized suite for xlarge TOSA model tests instead.
The reland of the DeepSeek layer tests can mark the memory-heavy cases as xlarge and route them through this suite. This also gives internal CI an explicit option for running any xlarge TOSA model coverage serially.
cc @digantdesai @freddan80 @per @zingo @oscarandersson8218 @mansnils @Sebastian-Larsson @robell @rascani