ComfyUI Extension: ComfyUI-ExtendIPAdapterClipVision
Run ComfyUI workflows without the setup
No installs, no CUDA version roulette, no GPU sitting idle on your bill. Bring a workflow and run it in the browser.
This extension provides two nodes to use with my experimental a/ip-adapter finetune for NoobAI-XL style transfer. a/Here is the counterpart extension for Reforge WebUI.
Looking for a different extension?
Custom Nodes (2)
README
Introduction
This extension provides two nodes to use with my experimental ip-adapter finetune for NoobAI-XL style transfer. Here is the counterpart extension for Reforge WebUI.
- Extend Clip Vision Input Size: Interpolate PE for the loaded clip-v model to make it able to accept images of a different size. One can use it to extend the original clip-v's input size from 224. (My finetune was trained on 448.)
- IPAdapter Advanced (Clip Size Aware): A modified version of the original ip-adapter node (credits to the author of ComfyUI-IPAdapter-Plus), instead of using a pre-defined value for image preprocessing, resizes the input to the actual clip-v input size.
Example Workflow
Run ComfyUI workflows without the setup
No installs, no CUDA version roulette, no GPU sitting idle on your bill. Bring a workflow and run it in the browser.