Problem/Motivation
gpt-image-1 and gpt-image-2 can take image input, we should make that possible.
Proposed resolution
Add ImageToImage operation type.
Remaining tasks
User interface changes
API changes
Data model changes
Issue fork ai_provider_openai-3593019
Show commands
Start within a Git clone of the project using the version control instructions.
Or, if you do not have SSH keys set up on git.drupalcode.org:
Comments
Comment #3
marcus_johansson commentedComment #4
marcus_johansson commentedComment #5
marcus_johansson commentedComment #6
arianraeesi commentedComment #7
abhisekmazumdarComment #8
abhisekmazumdarComment #9
abhisekmazumdarThis completely slipped my attention. Recently, I have been more focused on the GitLab work item, and I skipped checking the d.org assigned issues. Apologies for blocking this issue for days.
Comment #10
merilainen commentedComment #11
merilainen commentedReviewed and tested the changes, approved on my side. Made also a small code change requested by Marcus, so "assigning" the issue back to him now for final approval.
Comment #12
merilainen commentedComment #13
marcus_johansson commentedWe need to remove the `response_format` parameter in the api_defaults.yml, otherwise it looks good. Setting back to needs work and when done it can be set to RTBC.
Comment #14
merilainen commentedComment #15
merilainen commentedThe response_format removed from api_defaults.yml image_to_image section, marking as RTBC.
Comment #16
marcus_johansson commentedThanks, getting merged!