Invoke is a leading creative engine for Stable Diffusion models, empowering professionals, artists, and enthusiasts to generate and create visual media using the latest AI-driven technologies. The solution offers an industry leading WebUI, and serves as the foundation for multiple commercial products.
When creating a reference image (global or regional) we were not setting the CLIP vision model correctly. As a result, FLUX IP Adapters, which require the ViT-L model, were causing errors.
This situation is now handled.
Also, when graph building fails, we now display an error toast. Previously, there was no feedback at all, so it felt like the invoke button just did nothing.
Related Issues / Discussions
offline discussion
QA Instructions
This won't fix existing misconfigured IP Adapters, but it does prevent new ones from ending up with the wrong CLIP vision model selection.
Merge Plan
n/a
Checklist
[x] The PR has a short but descriptive title, suitable for a changelog
Summary
When creating a reference image (global or regional) we were not setting the CLIP vision model correctly. As a result, FLUX IP Adapters, which require the
ViT-Lmodel, were causing errors.This situation is now handled.
Also, when graph building fails, we now display an error toast. Previously, there was no feedback at all, so it felt like the invoke button just did nothing.
Related Issues / Discussions
offline discussion
QA Instructions
This won't fix existing misconfigured IP Adapters, but it does prevent new ones from ending up with the wrong CLIP vision model selection.
Merge Plan
n/a
Checklist