Qwen-image TXT2IMG, IMG2IMG, controlnet (canny, depth), lora manager, caption, tiled and SeedVR2 upscale. [DEPRECATED] - v1.4
A pretty simple workflow with a few common features. There are some important (more or less) notes inside the workflow.
This workflow is deprecated. I don't use Qwen-Image for a long time now. Krea2, Z-image, Klein are much faster than Qwen (without speed-up loras), give more effort. In particular, since release Qwen 2, that model has been abandoned quickly. So I want to save both my disk space and time.
The workflow is for legacy mode. I am not planning to make it useful or even test for Nodes 2.0 mode.
I tried Fun-Control ControlNet. For some reason, I always get a black image with any model and any settings, even in the official template.
Description
v1.1:
Fixed problems with the Input Parameters node after the comfyui-image-saver update.
v1.3:
Changed the resolution and resizing methods.
Added SeedVR2 upscaler.
Changed the metadata collection method.
v1.4:
Fixed the issue with collecting LORA metadata.