Announcement of high-quality image generation AI 'Stable Diffusion 3', capable of achieving highly accurate 'depiction of specified characters' and 'depiction of multiple subjects', which image generation AI is weak at.
Stability AI announced the image generation AI “ Stable Diffusion 3 ” on Friday, February 23, 2024. With Stable Diffusion 3, it is possible to perform operations that were difficult with conventional image generation AI, such as ``depicting specified characters without any discomfort in the generated image'' and ``describing multiple subjects in high definition.''
Stable Diffusion 3 — Stability AI Japan
Did you check out #stablediffusion3 that I introduced the other day? https://t.co/0I6DQuUzuQ
— Stability AI Japan (@StabilityAI_JP) February 25, 2024
In the thread below, we will introduce images actually generated with #stablediffusion3 .
The following is a 'cinematic photo of a red apple on a table in a classroom, on the blackboard are the words 'go big or go home' written in chalk' using Stable Diffusion 3. This image was generated with the prompt 'Go big or go home' (a movie-style photo with chalk written on a blackboard). Although there are differences in uppercase and lowercase letters, the words 'GO BIG OR go HOME' are written on the blackboard as instructed. Another point is that it is described as 'letters written in chalk' as instructed.
'a painting of an astronaut riding a pig wearing a tutu holding a pink umbrella, on the ground next to the pig is a robin bird wearing a top hat, in the corner are the words 'stable diffusion' A painting of an astronaut holding a pink umbrella, a robin wearing a top hat on the ground next to a pig, and the phrase ``stable diffusion'' written in the corner). The resulting image is below. In addition to being able to depict 'astronaut,' 'pig,' and 'robin' as instructed, the text string 'STABLE DIFFUSION' is written at the bottom left of the image. However, although the prompt says 'stable diffusion' in lowercase letters, I am concerned that it is written in uppercase letters in the generated image.
'Studio photograph closeup of a chameleon over a black background' looks like this. The area around the chameleon's face is clearly depicted, while the body is blurred, giving it the feel of being photographed with a macro lens.
Stable Diffusion 3 has multiple models with 8 million to 8 billion parameters. At the time of writing this article, it is in advance preview stage, and you can register for the waitlist from the link below.
SD 3 Waitlist — Stability AI
https://stability.ai/stablediffusion3
Related Posts:
