Stable Diffusion 3 API Released.

I'm a little worried about Stability AI's future with the round of layoffs and CEO departure. I hope whomever acquires them continues to keep the open source ethos.

The bg music is too loud... either you speak louder or just remove it totally.

please cover Hyper SD

I've used SD3 on google colab and it is very good indeed, unfortunately SD still doesn't know that a human being has 5 fingers 😭😭😭

ELLA did such a great job in prompt comprehension to the point that it’s difficult to see what SD3.0 is adding, if quality of example is what to go by.

Not free

So sad the dream of the community based models comes to a quick end... money making $$$

the best thing is: uncensored

Not free. Costs credits. F-all that garbage.

I have been enjoying SD2 w/a1111. Will something like that every be created for SD3 so i can run locally, for free, as much as I want?

Noone tell him that turtles are real.

I don't like it that SD3 currently looks bad compared to lykon's examples on his twitter page. I think it was a bad move to release a half-baked workflow version of SD3 that doesn't meet the aesthetic quality of MJ6. Looks like i'll be sticking with SD1.5 models for a while longer

Is Automatic1111 still the thing at this stage? or is time for me to move on do you think? I use all kinds of plugins etc.

"this is what turtles, uh, would've looked like if, uh, was kinda, half, semi, real"
turtles are the new pidgeons?

What's the point of sd3 if it as censored as mid journey and dalle 3, atleast Mj and dalle 3 has better image quality than sd3, i hope that a less censored version of sd3 will be available in the future.

hi there, is there a way of booking a call with you for consultation?

Have tested it SD3 and i am disappointed, prompt coherence in not even close to Dalle or ideogram... very sad...

haha great info, and the joke in the middle made my day!👏

interesting. the first time I watched this video the music wasn't distracting to me at all. I didn't even notice it that much. but liked that there was something in the background to listen

annoying background music. put an instrumental next time 😊

When they do the open release, will it be uncensored? I’ve just tried and for example “woman in bikini” fails because of content moderation

they are not gonna paywall everything are they?

The music was nice IMO, particularly when there weren't singers though. It's the additional "talking" I think that makes it particularly problematic, (plus it's in a similar register for a double whammy) that makes it tricky to understand/catch your voice alongside it. Do the Spanish uphold their ideals? Si'Bastion!

Why don't we have different open source models like LLM? For instance there are many architecture derived from transformers, like mistral, llama ... But for stable diffusion there are a lot of finetuned models but not new architectures. Do you know any other kind of architecture used to generate image with high quality like stable diffusion?

Why and how is all this free?

The music didn't bother me.

The background music is very loud and distracting.

Music? Man! You are not 13 years old boy!

Music too loud but ty for the vid.

Wouldve been nice to see more examples & how we can use it. Also its good to mention like half the leadership of Stability AI left during the last month or so due to their not actually being open.

Its semi open sourced...not fully.

do you have a tutorial about how to train stable diffusion to generate similar videos to the video you give it as a source? TY

I just wanna tell everybody in the comments that we have enough anime porn to last us a lifetime so if you guys wanna focus on some other type of art, that’d be great

❤

Still free or not?

Gunna be hard to get me off 1.5 with my 50 extensions, 100 trained models, and 3000 Loras.

Emad says SD3 can generate video as good as Sora. Did anyone try generating videos?

Now that we appear to have a functional text generation, I'm curious about the implications of copyright for the fonts in the training data. AI companies are already being sued by creators of text and images used in the training data… are foundries the next to jump into the fray?

Please no music. It sounds like another tab is open.

I am honestly salivating for the next few months.
Once the Community has had the time to Fine Tune SD3 And Develop the best practices to train the New Models and Loras.
Things are about to get really fun.

Stability: we MUST stop smut at all costs!

I hope my 6gb card can somehow run it, just to try it once.

half of the video and already i am shocked... is this a joke? stable diff 3 is this? even with instagram edits u can u do more... has to be a joke right?

"Thank you S. for a great video... It's great that you got to try SD3. When it comes to pricing for commercial use, which payment model are they talking about then, and how much? It would also be good to get a simplified explanation of what type of product use they envision is acceptable. Obviously, many have used SD because that model has been free and maintained a relatively good standard regardless of blood, boobs, and other personal nuances, which as you know are hard to even come close to with models like MJ."

So how do you use it?

The problem is not the music but the fact that there are lyrics.

I think text implementation will be better overall. The initial base model will always be the “start” and people will often overreact at the quality. As time goes on and with improvements and fine tuning, the forthcoming forms of SD3 will be better.

What are the technical differences between SD3, SD3 Turbo and Cascade?
Interesting video topic ?

SD could add an option for HUMAN FEEDBACK to continually improve, as with MJ ?

I couldn't watch the whole video because of the volume to the background music with lyrics. Too fatiguing.