[TRANSFORMATIONS] SDPAToPA add the model is stateful check #33002
+6
−0
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
CuriousPanCake
requested review from
Wovchena and
mryzhov
and removed request for
a team
Contributor
There was a problem hiding this comment.
Pull request overview
This PR adds a validation check to ensure that the SDPAToPagedAttention transformation is only executed on stateful models. The check verifies that the model contains variables before proceeding with the transformation, which previously existed in GenAI code but has been moved to the OpenVINO runtime.
Key Changes:
- Added an assertion to validate that the model is stateful (contains variables) before performing the SDPAToPagedAttention transformation
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Wovchena
approved these changes
Nov 24, 2025
CuriousPanCake
force-pushed
the
add_var_check_to_sdpa_to_pa
branch
from
693f903 to
0e6ce2d
Compare
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Outdated suggestions cannot be applied.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.
Details:
The SDPAToPA transformation can be executed only on a stateful model (to turn it into the stateless one). Previously, this check existed in the GenAI code, but since we've done transfer of some functionality to OV runtime, we need to add this check here.
Tickets:
Signed-off-by: Andrii Staikov [email protected]