Stability AI unveils powerful image generator SDXL 1.0, its rival to Midjourney

It is the largest open image model and is built on an architecture based on 3.5B parameters and a 6.6B parameter refiner.

Stability AI unveils rival to Midjourney

Those who have used SDXL 1.0 claim that the tool offers stunning images that could be better than Midjourney. (Image: Stability.AI)

Listen to this article

00:00

1x 1.5x 1.8x

It is the battle of image generation AI models it seems.Ever since Midjourney launched its generative AI-powered image generator, the world has witnessed a barrage of AI-generated images. This is perhaps why many tech leaders have infused the power of AI into their creative tools. Now, it seems there is a rival to Midjourney.

Stability AI has just released SDXL 1.0, an open model, and it is already seen as a giant leap in text-to-image generative AI models. The model is capable of generating high-quality images in any form or art style, including photorealistic images. The tool comes with enhanced ability to interpret simple language and accurately differentiate between homonyms. So far, it is the largest open image model and is built on an architecture based on 3.5B parameters and a 6.6B parameter refiner.

ICYMI | Pope in Balenciaga puffer jacket got you thinking? Midjourney v5 lets you create something similar

Those who have used SDXL 1.0 claim that the tool offers stunning images that could be better than Midjourney. SDXL 1.0 is the successor of SDXL 0.9. The company claims it to be the flagship model, and respondents have stated that they prefer it over other open models.

To test the creative abilities of SDXL 1.0, we used the prompt – “Yellow elephants flying in a meadow”. The tool not only produced an output but it also gave a slew of options. (Image via Clipdrop)

According to Stability AI, the latest model is well-tuned for vibrant and accurate colors, with better contrast, lighting, and shadows than the SDXL 0.9. “SDXL can generate concepts that are notoriously difficult for image models to render, such as hands and text or spatially arranged compositions (e.g., a woman in the background chasing a dog in the foreground),” the company said in its blog.

Also Read | Now artists sue AI image generation tools Stable Diffusion, Midjourney over copyright

SDXL 1.0 requires only a few words to create complex and aesthetically stunning images. Besides, users no longer need to use qualifier terms like “masterpiece” to get high-quality images. It can also understand the differences between concepts like “The Red Square” (a famous place) and a “red square” (a shape). The tool can be accessed from https://clipdrop.co/stable-diffusion

Tags:
artificial intelligence