flux-3-image-klein-comfy: Make and edit images in ComfyUI (5.6B)
Added · Updated
Made by Black Forest Labs
View profile on Hugging Face (opens in a new tab)Create images from text or edit pictures using reference images inside ComfyUI, for people who make images and run ComfyUI themselves.
Model details
Original model black-forest-labs/flux-3-image-klein-comfy
Input Image Output Image
Ways to use it Run locally
What it does
You give it a written prompt and up to 10 reference images, and it returns one image. A separate prompt upsampler first expands a short request into a full scene description, which the model needs for good results. You can place items with boxes on a 0 to 1000 grid, and it can draw text inside images.
How to use it
Put the files into ComfyUI's diffusion models, text encoders and vae folders, then run the provided workflows, which use the int8 prompt upsampler by default. Use the distilled checkpoint for making and editing images; the Base checkpoint is meant for fine-tuning and research.
Common use cases
- Type a short idea for a poster, let the upsampler expand it, and get one finished image that includes the poster text.
- Give a product photo as a reference and ask in the prompt to recolor or remove an item, getting an edited image back.
- Combine several reference images into one new scene, using boxes in the prompt to show where each element goes.
Can I use this?
Requirements First accept the FLUX Kommunity License on the gated Hugging Face page. Then pick file versions: the distilled model is 11.1 GB in BF16 or 5.8 GB in int8, with smaller-format options listed.
Caveat Boxes guide where things go and how big they are, but they are not exact masks. GPU and memory requirements are not documented. Commercial use depends on the license's Qualifying User terms.
Worth knowing License allows non-commercial use, with a commercial output exception only for Qualifying Users under its terms. Short prompts give generic results without a prompt upsampler. Requires accepting the license on Hugging Face (free; access is granted automatically).
Why it's here
- 2,722 people have liked it on Hugging Face.
- It was downloaded 1,167,959 times in the last 30 days.
Numbers as of 8 November 2026, from BuildTube’s latest daily refresh.
Behind it
View model pageBuildTube has not run or verified this project. Everything above is written from what the creator published.
Made this?
Quick questions
- What is flux-3-image-klein-comfy?
- You give it a written prompt and up to 10 reference images, and it returns one image.
- What can I use flux-3-image-klein-comfy for?
- Type a short idea for a poster, let the upsampler expand it, and get one finished image that includes the poster text; Give a product photo as a reference and ask in the prompt to recolor or remove an item, getting an edited image back; Combine several reference images into one new scene, using boxes in the prompt to show where each element goes
- Can I run flux-3-image-klein-comfy on my computer?
- Then pick file versions: the distilled model is 11.1 GB in BF16 or 5.8 GB in int8, with smaller-format options listed. GPU and memory requirements are not documented.
- What do I need to use flux-3-image-klein-comfy?
- ComfyUI, accepting the FLUX Kommunity License on the gated Hugging Face page, and disk space for several large files.
- What license does flux-3-image-klein-comfy use?
- License allows non-commercial use, with a commercial output exception only for Qualifying Users under its terms. Commercial use depends on the license's Qualifying User terms.
More from this week
- P-H3-Image-Comfy Make and edit 8-megapixel images in ComfyUI
Turn short text requests into 8-megapixel images and edit pictures on your own graphics card.
- Idol Turn one picture of an object into a 3D Gaussian splat
Idol turns one photo, render or generated image of an object into a 3D Gaussian splat file.
- agent-desktop Let AI agents read and click desktop apps
For developers building computer-use agents: agent-desktop lets an AI agent read and operate desktop apps on Mac, Windows or Linux from a command line.
Explore by topic
Post this
Download share imageFor makers
Made this? Show it’s on BuildTube — add the badge to your README or embed the card on your site.
README badge (Markdown)
[](https://buildtube.app/p/hf-black-forest-labs-flux-3-image-klein-comfy/)Embed card (HTML)
<iframe src="https://buildtube.app/embed/hf-black-forest-labs-flux-3-image-klein-comfy/" width="560" height="240" style="border:0;max-width:100%" loading="lazy" title="flux-3-image-klein-comfy on BuildTube"></iframe>