restore SAM3 interactive fast path for repeated clicks#2480
Merged
Conversation
… segment_image is called
grzegorz-roboflow
requested review from
PawelPeczek-Roboflow,
dkosowski87,
hansent,
probicheaux,
rafel-roboflow and
yeldarby
as code owners
grzegorz-roboflow
changed the title
| self.task_type = "unsupervised-segmentation" | ||
|
|
||
| # Keep interactive embeddings/logits on GPU to match legacy click latency; | ||
| # make CPU spill configurable if memory pressure shows up. |
Collaborator
There was a problem hiding this comment.
voting for making this env-variable based with default True and adjusted on serving
PawelPeczek-Roboflow
approved these changes
Jun 18, 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What does this PR do?
This change restores the SAM3 interactive fast path for repeated clicks. When
image_id is provided, the adapter now tries cached embeddings before decoding/
preprocessing the image, falling back to image input only on cache miss.
It also keeps SAM3 interactive embedding/logit caches on GPU to match legacy click
latency behavior, and adds unit coverage for cache hit, cache miss fallback, error
propagation, and cache placement.
Type of Change
Testing
Checklist
Additional Context