AI generator model in few words
AI models are basically advanced statistical models based on the principle of machine or deep learning. In our case their inputs are photos from all over the internet, which are marked with what exactly they present. Based on this, the neural network learns to "understand" how things look. Thanks to that, based on the textual description, the neural network can generate statistical variation with surprising accuracy.
You can learn more, if you are a newbie.
How it works?
- Describe the type, style, content of image you want to create in a few words
- AI image generator will create a unique image based on your description
- You can save final image to your storage and use it "however you like"
Ok, enough theory, let's try it..
List of AI art generators
This list will be constantly updated.
DALL-E
Creates images from text captions for a wide range of concepts expressible in natural language.
- DALL-E demo
- Price: free, open source
- Results quantity: ★ ★ ★ ★ ★ ★ ★ ★ ★ ★
The first generation of the model, which subsequently became the best in version 2. The predecessor of DALL-E 2 - which is of course temporary best of all models.
Example results
DALL-E 2
New AI system that can create realistic images and art from a description in natural language.
- DALL-E 2 demo
- Price: paid
- Results quantity: ★ ★ ★ ★ ★ ★ ★ ★ ★ ★
The results of the model are currently the best of all in terms of realism and fidelity. This technology is currently only available to a selected group of users due to possible obsolescence.
Example results
Midjourney
Midjourney is an independent research lab exploring new mediums of thought and expanding the imaginative powers of the human species.
- Midjourney demo
- Price: free and then paid
- Results quantity: ★ ★ ★ ★ ★ ★ ★ ★ ★ ★
I feel like this model was trained primarily on sci-fi and fantasy art works. At the same time, the results are suspiciously beautifully lit so that everything is pleasing to the eye. When you write a simple text and the result is so beautiful, it is almost irresistible to try new and new combinations for which you ultimately have to pay. I think it's tuned to make people like the results and pay big. It becomes an addiction, at least until you get enough of your "creativity".
Example results
Stable Diffusion
Stable Diffusion is a text-to-image model that will empower billions of people to create stunning art within seconds. It is a breakthrough in speed and quality meaning that it can run on consumer GPUs.
- Stable Diffusion demo
- Price: open source
- Results quantity: ★ ★ ★ ★ ★ ★ ★ ★ ★ ★
It is an open source solution that you can try through third-party front-ends and also locally on your computer.
Local apps to use Stable Diffusion on local computer
- Visions of chaos - multiplatform UI
Other applications using Stable Diffusion model:
- Art Breeder: https://www.artbreeder.com/beta/collage