I was using stable diffusion a lot previously, but haven’t really touched it in the past several months. I was wondering what interfaces people are using these days?
Automatic1111 still seems to be popular, and that’s the one I am most familiar with. I know there are some others now though, like comfy, and I guess maybe invokeAI is still going?
Automatic1111 is the second one I tried and I never left. It’s the best.
Automatic1111’s Stable Diffusion WebUI is hard to give up, with how many features it has that are missing in other frontends. I use anapnoe’s fork for a slightly better UI. I would use vladmandic’s fork, but some of the changes have caused issues for my particular setup.
Is no-one here running ComfyUI? It’s one of my favorite UI’s as it’s completely node based and extensive! Has everything auto1111 has and even more! EDIT: It’s not really everything, but almost!
I’m also using ComfyUI. It just has the ability to do so much more than something like Automatic1111, even if it is missing a couple of features. For example, I have several workflows that do incremental changes to a photo, changing the prompt halfway through the generation, or even upscaling halfway through the generation.
I can’t really imagine going back, unless there is some killer feature that Comfy is missing.
Automatic1111
Draw Things and InvokeAI on my MacBook Pro m1 pro
SD.next (Vladmandic)
I use DiffusionBee on Mac. It’s simple and good enough for my needs
I’m stuck with an AMD card for other purposes so I pretty much have to use the DirectML fork which is okay, but it’s very slow and despite having 12gb of video ram, I still get the out of ram messages all the time. But, hopefully some progress will be made somewhat soon on those cards.
But it is fun, that’s for sure.
I mainly use the UI-UX fork of AUTOMATIC1111.
I think Automatic is already great, but you should at least mention Easy Diffusion. Also very easy to use.
i used to use invoke ai,. invoke ai just realeased version 3 beta. i will wait for a few more betas or rc for 3 to use it again.
These days I usually use MakeayoThe porn industry is doomed!
Edit: well I think the title of this post changed or my comment ended up on the wrong post so my comment is irrelevant lol
Not for videos, we are still quite a way from that yet.
yeah it was a joke. Doesn’t look like it resonated well lol.
Haha it won’t be a joke next week when the new text-to-video model comes out
With how unstable (lol) txt2img is, I don’t believe a stable enough for porn txt2video model is coming soon.
I would agree, but the rate of innovation in AI is so unpredictable that it could go either way.
I don’t really agree.
Recent AI inovations are pretty modest and use the innovation of raw fucking power to achieve goals.
Gpt4 uses 230B parameters, whereas to run a 7B LLM you need 16gb of vram already, and llms are o(n²) in complexity in terms of parameters, I’ll let you do the maths
Stable diffusion (latent diffusion to be more precise) is about the same, the initial training required billions of teraflop, while it was relatively cheap (100k$), it still rides on modern GPU technology .
👀
