This model is named Cinematic Diffusion. It has been trained using Stable Diffusion 2.1 (512px) to generate cinematic images. To utilize it, you must include the keyword "syberart" at the beginning of your prompt.
This model performs best in the 16:9 aspect ratio, although it can also produce good results in a square format. (You can use 906x512; if you have duplicate problems you can try 968x512, 872x512, 856x512, 784x512).
Cinematic Diffusionis well-suited for creating 'screengrabs' of movies from a variety of genres including historical, sci-fi, fantasy, spy, horror, western, mystery, comedy, superhero, anime, and more. It can also be used for creating realistic portraits, landscapes, and other types of images.
Cinematic Diffusion was created by Basunat (Support: https://ko-fi.com/basunat Shop: basunat.redbubble.com) and uploaded here with consent.
Settings: For optimal results, it is recommended to use the default settings on Automatic 1111. It works on SD 1.5 and SD 2.1 (512px)
Description
V1 SD 2.1 (512px)
FAQ
Comments (4)
i think that is trained with some south asian action movie. The characters are somehow in the same range. Is good though for some foggy, dark scenes.
Have you tried it or are you just talking from the example images? It has not been trained using images from any movie, but images previously generated with SD. If you change the prompt you can create very different images, characters and brighter scenes if that is what you describe in your prompt. It is a pity that I cannot upload images here to demonstrate the very different scenes that can be created with this model. You can also use the negative prompt to remove a subject that you don't want to appear in the results for more variety.
Thank you this model is FIRRRRRE
can we get a safetensor version, please?
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.