Add configurable VLM image preprocessing strategy#108
Merged
Conversation
Support three image preprocessing strategies for VLM vision encoders: - stretch: resize directly to target (default, backward compatible) - center_crop: shortest-edge resize then center crop (for CLIP-based models) - pad: longest-edge resize with zero-padding (preserves geometry) The strategy is declared in metadata.json and inferred from the model at export time. Runtime CLI overrides via --image-strategy. Also adds --image-info flag to optionally inject original image dimensions into the text prompt (useful for models trained with resolution awareness). Closes apple#100.
stikves
marked this pull request as ready for review
July 16, 2026 18:44
stikves
requested review from
alejandro-isaza,
carinapeng,
kevchengcodes and
tjia1818
July 16, 2026 18:47
kevchengcodes
approved these changes
Jul 16, 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Support three image preprocessing strategies for VLM vision encoders:
The strategy is declared in metadata.json and inferred from the model at export time. Runtime CLI overrides via --image-strategy.
Also adds --image-info flag to optionally inject original image dimensions into the text prompt (useful for models trained with resolution awareness).
Closes #100.