ChatGPT recently allowed the ability for all users to take a photo they have and convert it into Studio Ghibli style aesthetics. It works well but quite slow with free plan but paid version is faster.
I’m sure many of the people (tech savvy) have been wondering how I can do this. Well its quite easy using Image to Image processing in ComfyUI. You can use a fast/turbo SDXL (Stable Diffusion XL) model with a Studio Ghibli LoRA to generate the desired result.
I found a workflow on my computer that I had downloaded a while back (sorry if I cannot credit original maker of the workflow) and started to modify it to get the results that resemble the input image.
If you are new to ComfyUI then check out my elaborate tutorial on ComfyUI in this post Getting Started with ComfyUI in 2025. Here I go through most of the main things you need to understand to get started with ComfyUI.
Once you are in ComfyUI simply import my workflow (download below) and Install Missing Custom Nodes. Then you need to download these items if you don’t have them already.
I’ve added to the workflow Florence2Model which helps with interpreting the image you load and supplying that as a prompt, this model is downloaded automatically first time you run the workflow and saved locally for future usage. So no need to download it again. If you already have this node pack installed then it won’t download the model. You can also change or choose a different Florence2 model or another alternative.
Since creating the workflow and writing this post I added a colour correction node which improves the vibrancy of the resulting image. The output image is only a Preview so you can replace it with Save Image node if you want to save them.
If you want to batch process images then consider using my WW Image Batch Loader node which should be installed already when you install the missing custom nodes. This can interrogate a batch folder and load images one by one to process in your workflow.
Download
Download the workflow which comes in a zip file containing two formats in PNG and JSON. Simply uncompress the zip file and drag any of these files into ComfyUI to load the workflow.
SDXL Studio Ghibli Style Workflow (2 downloads )
Results
Here are some of the results I was able to get with this workflow.
If you’d like to support our site please consider buying us a Ko-fi, grab a product or subscribe. Need a faster GPU, get access to fastest GPUs for less than $1 per hour with RunPod.io
