- More cool models.
ℹ️ Hyper Realism 3 training included 54,000 images. A lot of historical images, realistic photography styles, style per decade, misc styles, and actresses without focusing on faces.
ℹ️ Trigger words:
2000s, 1990s, 1980s, 1970s, 1960s, 1950s, 1940s, aleksei gilev style, andrew vasiliev style, elenok ella style, ivan kovalev style, jiri tulach style, komrant style, marie dashkova style, olga boyko style, digital art, interior design, macro photography, polaroid style, sucidegirl, wild natureDescription
ℹ️ This model is a general-purpose realistic model.
FAQ
Comments (5)
Just by looking at the outputs alone, it's really nice.
Thank you very much.
You forgot to add: awsome, super super super realistic or more realistic then real. I often wonder what kind of drugs people are using calling there crappy illustrations, realistic or even art, lol
Hi @seedstyler,
Even the most realistic Stable Diffusion models were trained with a large number of illustrations and cartoons, which help models to achieve a higher level of abstraction to represent various subjects. It's part of the core training of all models, provided to us by Stability AI. It only costs a few million dollars to train a model from nothing, so you can do it if you want.
Not all examples that I've used are 100% super realistic, you can prompt for more realistic images if you want. It's part of the fun of using Stable Diffusion.
My model Hyper Realism 1.2 is a model with a base style tilted toward realism, and with skilled prompting, you will be able to achieve amazingly vivid and realistic results.
At these moments, I'm already running a training of Hyper Realism 2.0. I hope to hear your opinion, whether you like it better than 1.2, when I release it.
I hope you are having a wonderful day, and thank you for being such a charming fellow.
@alexds9 This is a very classy response, one which deserves my praise. :)
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.