Stability AI recently released open model SDXL 1.0 the next iteration in text-to-image generation models. The full version of SDXL has been improved to be the world’s best open image generation model compared to its previous research-only release of SDXL 0.9.

After weeks of extensive testing, SDXL 1.0 was the best image model preferred over other models by users input:

SDXL 1.0 is the flagship image model from Stability AI and the best open model for image generation. We’ve tested it against various other models, and the results are conclusive – people prefer images generated by SDXL 1.0 over other open models. This research results from weeks of preference data captured from generations of experimental models on our Discord and from external testing

SDXL is considered the best open model for photorealism and creates images of high quality in virtually any art style. SDXL really excels in replicating vibrant and accurate colors:

SDXL 1.0 is particularly well-tuned for vibrant and accurate colors, with better contrast, lighting, and shadows than its predecessor, all in native 1024×1024 resolution.

SDXL uses only a few words to create “complex, detailed, and aesthetically pleasing images.” SDXL can also distinguish between similar words that have different meanings.

SDXL 1.0 is the largest open image model and has “one of the largest parameter counts of any open access image model, built on an innovative new architecture composed of a 3.5B parameter base model and a 6.6B parameter refiner.”

See also  PWC Predicts Six AI Priorities for 2019

A two-stage architecture, the base model generates (noisy) latents and a refinement model specialized for the final denoising steps result in “image generation without compromising on speed or requiring excess compute resources.”

chipmunk nerd 600

Mage’s AI text to image generator output

Mage recently announced its introduction of SDXL to Mage’s AI text to image generator which is free and unlimited for all users.