I would like to experiment with image editing. Idk, brush up my vacation photos, remove some of the other tourists, make it look like I've been somewhere I haven't... Swap my face onto someone or change my clothes, change the weather/sky... Draw me as a character in some artist's style... If it works well enough, take some of the blurry images from my potato phone camera from 15 years ago and clean them up... The stuff I've seen people on YouTube do with AI.
What's a good and recent model or "editor" to do that? Open-weights please and preferably something I can just use straight away without fiddling 3 hours with some ComfyUI nodes. Bonus points for being efficient and running on my laptop 😅
And is there some Free Software UI / software to make it easy? I mean I'm perfectly able to learn how to provide it with stencils(maybe they don't need that anymore in 2026?) and whatever stuff it needs as input, run it on the commandline... But maybe there's a nice interface like an image editor?
Edit: I forgot: I mentioned different use-cases... Seems to me there are different models for like regular image editing (which gets one image and a prompt as input) and models which get two images as input? Like a picture, and another picture and the instruction to draw this person onto that picture?
9 Comments
hendrik@palaver.p3x.de · 7 pts · 2d
Why all the downvotes? General AI hate drive-by nastiness? Or did I do something wrong?
Smorty@lemmy.blahaj.zone · 7 pts · 2d
i think so, yes. i believe that FOSAI is mooostly about language and embedding models and such and less about image, video and audio gen.
and also yesssss, the community is on lemmy world, so many people will see the post, see image gen and dislike. that's just how it be.
i don't got an answer to your question, sorry. am not an image gen kinda person.
hendrik@palaver.p3x.de · 3 pts · 2d
Thanks for your kind answer. Maybe I'll ask in the stable-diffusion community in case I don't get any answers. I just figured I'm massively into open source and maybe I'll find some similar-minded people here. And/or people who don't do it all day as a hobby and have a beefy graphics card and super elaborate setups... Because I don't, I just want it as a tool 😅
Hamartiogonic@sopuli.xyz · 6 pts · 1d
Because Lemmy.
This is the corner of the internet where you’ll find people who hate AI with a burning passion. Apparently, some of them were scrolling local or all, and stumbled upon your post. Some people just downvote without even checking which community the post is in.
EnsignWashout@startrek.website · 6 pts · 2d
Stable Diffusion is open source and at least has options for all of that in the web UI (which you can host locally).
I wouldn't bet on successful runs on a latop, though. I have a powerful desktop gaming rig, and I still get OOM (out of memory) errors trying to do things with it.
Of course, there's definitely a crazy learning curve for optimizing this stuff. So by "can't work on a laptop" what I really mean is that I couldn't get it working on my laptop, when I tried.
hendrik@palaver.p3x.de · 2 pts · 2d
Thanks but I guess regular Stable Diffusion is a bit too old for my taste. Seems we have better, more efficient models as of today and they're way easier to prompt, can do text, the right amount of fingers... And all of that requires constant fiddling with the earlier models.
keepthepace@tarte.nuage-libre.fr · 5 pts · 2d
A while ago automatic111 was the way to go for easy image manipulation with diffusion models but people seem to have largely migrated to Comfy for, well anything, for better or worse.
I finally bit the bullet when I wanted to try krea2. And I recommend it. You will be doing things in less than 3 hours if you know what is a latent and a clip model. Start from a working workflow close to what you want to do and fiddle from there.
If you want models recommendation, you should post your VRAM budget.
For image editing there are basically two ways of doing it:
models especially trained o modify an image from a text prompt like qwen image. Some accept several image references and you can do things like "person from picture 1 in the setting from picture 2 and smoking a cigarette"
models that do inpainting, which is most image generation models, it is the tooling around them that does the fiddling: feed them the base image and the mask you want to fill, as well as a prompt and it will fill the mask with what you proposed.
Comfy has a masks editor and allows you easily to transform a pipeline into an app.
I kinda understand why everyone moved to there because the field is moving fast and it is much easier to write nodes for a new part that re-designing a whole workflow like automatic111 used to do.
BrianTheeBiscuiteer@lemmy.world · 3 pts · 1d
I started with Auto1111, then moved to comfy, then started bouncing between tools including Invoke. Auto1111 feels antiquated (mostly due to it's setup process), Comfy has been good but there's so many nodes it can take an hour or more to setup a workflow I downloaded (plus many nodes are using the cloud which I want to avoid). Invoke has been nice for the few days I've used it and the biggest negative so far is dealing with the layering biz in the canvas.
How does krea2 compare?
hendrik@palaver.p3x.de · 1 pts · 2d
Yeah, my VRAM budged is a bit of an issue. I don't own any GPU 😆 So I'm basically fine with every recommendation. Either it's some "Turbo" model doing just 4 steps, I'll be patient run it on CPU and just wait 20min for my image being edited... Or I'm gonna have to rent a bit of compute on a service like Runpod. I can do both. And cloud compute can do 80GB of VRAM or whatever I put in. It's just that I'd like to know what to try beforehand, because I then get billed by the minute.