Update to Ideogram4 JSON prompt tool
https://preview.redd.it/dgj9neords5h1.png?width=1364&format=png&auto=webp&s=c9f252cfea67f6b5d37218621ef75306d604b641
A couple days ago I made a small JSON prompt builder for IdeogramV4 which some people found helpful, so posting a small update here.
Main changes are PNG metadata import, photo/art\_style mode switching, and larger image size ranges.
# PNG import
You can drag an image generated in Comfy into the tool's window, and if it contains a JSON prompt in the correct format, it will load the bounding boxes and JSON parameters used to create the image. Imported images do not need to be made with this tool, any image generated with ComfyUI with a valid JSON prompt should work.
This should help with quicker iteration
# Photo/art_style mode switching
The [official Ideogram4 prompting guide](https://github.com/ideogram-oss/ideogram4/blob/main/docs/prompting.md) lists two modes for JSON prompt structure, one for photographic and one for non-photographic images. Both are now supported in the tool.
(I HIGHLY recommend reading the prompting guide, there are lots of useful tips to get the most out of this model)
# Image size range
Ideogram 4 performs surprisingly well with large images, so the image size slider ranges have been increased up to 4096.
The github is [https://github.com/d-daley/ideogram4-editor](https://github.com/d-daley/ideogram4-editor) if you want to clone the tool locally, or you can access it in the browser here: [https://d-daley.github.io/ideogram4-editor/](https://d-daley.github.io/ideogram4-editor/)
https://redd.it/1tz1uaq
@rStableDiffusion
https://preview.redd.it/dgj9neords5h1.png?width=1364&format=png&auto=webp&s=c9f252cfea67f6b5d37218621ef75306d604b641
A couple days ago I made a small JSON prompt builder for IdeogramV4 which some people found helpful, so posting a small update here.
Main changes are PNG metadata import, photo/art\_style mode switching, and larger image size ranges.
# PNG import
You can drag an image generated in Comfy into the tool's window, and if it contains a JSON prompt in the correct format, it will load the bounding boxes and JSON parameters used to create the image. Imported images do not need to be made with this tool, any image generated with ComfyUI with a valid JSON prompt should work.
This should help with quicker iteration
# Photo/art_style mode switching
The [official Ideogram4 prompting guide](https://github.com/ideogram-oss/ideogram4/blob/main/docs/prompting.md) lists two modes for JSON prompt structure, one for photographic and one for non-photographic images. Both are now supported in the tool.
(I HIGHLY recommend reading the prompting guide, there are lots of useful tips to get the most out of this model)
# Image size range
Ideogram 4 performs surprisingly well with large images, so the image size slider ranges have been increased up to 4096.
The github is [https://github.com/d-daley/ideogram4-editor](https://github.com/d-daley/ideogram4-editor) if you want to clone the tool locally, or you can access it in the browser here: [https://d-daley.github.io/ideogram4-editor/](https://d-daley.github.io/ideogram4-editor/)
https://redd.it/1tz1uaq
@rStableDiffusion
