51–60 of 157 posts
Earlier quoted context omitted.
Normal SDXL requires 1024x1024 output or the quality degrades significantly.
Hi Max! Thanks for all the tuts! Small correction - SDXL wants ~1 megapixel resolutions at a variety of aspect ratios. https://github.com/lllyasviel/Fooocus/issues/24
Earlier quoted context omitted.
I've done a bit of fiddling around with it and definitely holding back judgement for now, seems like the 1 and 2 step images are WAY more coherent than LCM, but the images are kinda trash for any kind of prompt complexity so you start to have to use more steps, and since the individual steps take the same amount of time (I think there's a specific sampler for this which may be faster & better?) by the time you start…
Are you finding dpm++ 3M SDE better than dpm++ 2M SDE in sdxl? Afaik the second order (2M) version is the recommended one to use for guided sampling vs the 3rd order one. From here: https://huggingface.co/docs/diffusers/v0.23.1/en/api/schedul... > It is recommended to set solver_order to 2 for guide sampling, and solver_order=3 for unconditional sampling.
Which is what it's meant for, I suppose.
Noncommercial use - aside from being one of my licensing pet peeves - seems to indicate that the money is drying up. My guess is that the investors over at Stability are tired of subsidizing the part of the generative AI market that OpenAI refuses to touch[0]. The thing is, I'm not entirely sure there's a paying portion of the market? Yes, I've heard of people paying for ChatGPT because it answers programming questio…
Stability and Midjourney are also making money, but it’s largely with amateurs and people prototyping content for their own creations. A lot of single person Indy game developers are using these tools to generate assets, or at minimum first pass assets. I think a lot of media companies are producing the art for their articles or news letters etc using these tools. Whether this is enough, I don’t know.
Earlier quoted context omitted.
I've done a bit of fiddling around with it and definitely holding back judgement for now, seems like the 1 and 2 step images are WAY more coherent than LCM, but the images are kinda trash for any kind of prompt complexity so you start to have to use more steps, and since the individual steps take the same amount of time (I think there's a specific sampler for this which may be faster & better?) by the time you start…
Are you finding dpm++ 3M SDE better than dpm++ 2M SDE in sdxl? Afaik the second order (2M) version is the recommended one to use for guided sampling vs the 3rd order one. From here: https://huggingface.co/docs/diffusers/v0.23.1/en/api/schedul... > It is recommended to set solver_order to 2 for guide sampling, and solver_order=3 for unconditional sampling.
Earlier quoted context omitted.
They have a web demo here: https://clipdrop.co/stable-diffusion-turbo
...which requires you to sign in. That nice little text box invites you until you actually click to enter some text and get a registration box thrust at you People that design a UX where the user tricked into a registration 'ambush' need to be punched in the face.
UX designers doing their job and programmers protecting their free product from being abused are not the people deserving abuse here.
The license is non-commercial, but: > For clarity, Derivative Works do not include the output of any Model. https://huggingface.co/stabilityai/sdxl-turbo/blob/main/LICE... Doesn't that mean that generated images from it should be fine for commercial?
> On an A100, SDXL Turbo generates a 512x512 image in 207ms (prompt encoding + a single denoising step + decoding, fp16), where 67ms are accounted for by a single UNet forward evaluation.
Okay... so what part of this is real time? 207ms is 4.8Hz. 67ms is 14.9Hz. Isn't "real time" in graphics considered to be at least 30Hz (33ms)? And by today's standards at minimum 60Hz (16ms) if not 144Hz (7ms)? I'm lost at what part of this is real time? I'm not sure it even would get there with an H100. Maybe an H100 and everything is TensorRT?
> A Real-Time Text-to-Image Generation Model > On an A100, SDXL Turbo generates a 512x512 image in 207ms (prompt encoding + a single denoising step + decoding, fp16), where 67ms are accounted for by a single UNet forward evaluation. Okay... so what part of this is real time? 207ms is 4.8Hz. 67ms is 14.9Hz. Isn't "real time" in graphics considered to be at least 30Hz (33ms)? And by today's standards at minimum 60Hz (1…
Earlier quoted context omitted.
Hi Max! Thanks for all the tuts! Small correction - SDXL wants ~1 megapixel resolutions at a variety of aspect ratios. https://github.com/lllyasviel/Fooocus/issues/24
Fair point, although in my experience SDXL still isn't great at non-square ratios. I end up just cropping them to the ratio I want.
Noncommercial use - aside from being one of my licensing pet peeves - seems to indicate that the money is drying up. My guess is that the investors over at Stability are tired of subsidizing the part of the generative AI market that OpenAI refuses to touch[0]. The thing is, I'm not entirely sure there's a paying portion of the market? Yes, I've heard of people paying for ChatGPT because it answers programming questio…
The US breaking Google up also won't end the monopoly money. More likely one of the children will spin out with an even better margin business.