I tested all possible combination for Stable Diffusion 3.5 Turbo

This is again a very useful, interesting test! I personally would be more interested in the Large version, because it focuses on image quality rather than speed.

[@CodeCraftersCorner]

Thanks! Yes, I am doing the large version but it's taking longer

[@sudabadri7051]

Amazing work mate

[@CodeCraftersCorner]

Thanks a lot!

[@CalladsEssence]

Thank you Sharvin for this comparison!

[@CodeCraftersCorner]

Thank you very much! Thanks for the support!

[@pawelthe1606]

For me the best settings tested for realism (portrait) are: Model Sampling SD3 shift on 2.00, steps on 4, cfg on 1,sampler - euler, scheduler-beta. Time for generation with amazing results about 15 seconds.

[@CodeCraftersCorner]

Thanks for sharing! Will try it.

[@PixelsVerwisselaar]

Nice thanks 🙌

[@CodeCraftersCorner]

Thank you too

[@pika9985]

would you do the same test but for realistic picture result ? in a new video

[@pawelthe1606]

Try this settings: Model Sampling SD3 shift on 2.00, steps on 4, cfg on 1,sampler - euler, scheduler-beta. Time for generation with amazing results about 15 seconds.

[@pika9985]

@@pawelthe1606 thanks and unfortunately now i'm able to use cpu interference , os either turbo sd35 or waiting for a similar lightning 4step lora , then i stuck at cfg 1 , so i ask about what best scheduler and sampler for realistic output

[@CodeCraftersCorner]

Yes, I planned to experiment with realistic pictures too.

[@pedrogorilla483]

No, please do it in the same video 😜

[@Kaoru8168]

I use the deis sampler for everything it works for 3.5 too

[@CodeCraftersCorner]

Thanks for sharing.

[@zephilde]

Hi! Did you try the same with more steps or on non turbo model ?

[@CodeCraftersCorner]

Hello! Yes, I am doing the large version but it's taking longer. I will post the results once completed.

[@zoom1972idi]

Отличное видео, спасибочки. Можно вашу схему для Comfy Ui?

[@CodeCraftersCorner]

Hello, yes. You can use the images from the GitHub repository (link in description). You can drag and drop the image onto ComfyUI to load the workflow.