invoke-ai / invoke-ai/InvokeAI

[enhancement]: Metadata should support multi-image IP adapters

Open
#5,776 0 comments 0 reactions 0 assignees View on GitHub
enhancement
Dominant language
Python
Stars
28.2k
Forks
3k
Avg merge
6d 5h
Merged PRs (30d)
19

Description

### Is there an existing issue for this?

- [X] I have searched the existing issues

### Contact Details

skunkworxdark

### What should this feature add?

When saving metadata in a workflow it should support multi-image IP adapters.

In a workflow, the IP-Adapter node can take a collection of images. However, when you save the metadata for this it doesn't work. By that I mean nothing gets added to the metadata under "ipAdapters":

### Alternatives

_No response_

### Additional Content

In a #dev-chat discord discussion ([1209966022837280890](https://discord.com/channels/1020123559063990373/1049495067846524939/1209966022837280890)) with @psychedelicious it he mentioned that ` "ip adapter metadata is validated using a separate frontend schema" `

Also, I found that in `metadata.py` the `class IPAdapterMetadataField(BaseModel):` has a definition for the image field that does not allow lists.

Finally, the frontend linear UI doesn't provide a way to have multi-image ip-adapters so any modification to the metadata data shouldn't break the ability to restore settings from metadata for ip-adapters.

Contributor guide

No contributing guide indexed for this repository

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.