On June 25, Stability AI announced the official launch of SDXL 0.9, their text-generation image model. This version boasts improved image content and composition details and can operate on consumer-grade GPUs.
SDXL 0.9 reportedly has the largest number of parameters among all open-source image models, with a base model of 3.5 billion parameters and an additional model of 6.6 billion parameters.
The working principle of SDXL revolves around using the basic model to create rough details, and then refining the generated pictures with the additional model. If you have experience with Stable Diffusion, you may have noticed this as a progressive working process.
Two CLIP models are used in SDXL 0.9, including OpenCLIP vitg/14, the biggest OpenCLIP model to date. With this model, Stable Diffusion can create lifelike images with greater depth and higher resolution, according to Stability AI.
Stability AI also announced that the SDXL team will release a research blog that details the SDXL 0.9 model specifications and more parameter details. It is expected that the model will be open-sourced on GitHub as version 1.0 in July.
So guys, if you liked this post and wish to receive more tech stuff delivered daily, don’t forget to subscribe to the Inspire2Rise newsletter to obtain more timely tech news, updates and more!
Keep visiting for more such excellent posts, internet tips, and gadget reviews, and remember we cover, “Everything under the Sun!”