SDXL 0.9: A Leap Forward in AI Image Generation

4 min readJul 3, 2023


If you are looking for a way to create stunning visuals with just a few words, you might want to check out SDXL 0.9, the latest development in the Stable Diffusion text-to-image suite of models by Stability AI.

SDXL 0.9 is a game-changer for creative applications of generative AI imagery. It can produce hyper-realistic images for various media, such as films, television, music, and instructional videos, as well as offer innovative solutions for design and industrial purposes. SDXL 0.9 sets a new standard for real-world uses of AI imagery.

What is SDXL 0.9?

SDXL 0.9 is a text-to-image model that can generate high-quality images from natural language prompts. It is based on the Stable Diffusion framework, which uses a diffusion process to gradually refine an image from noise to the desired output.

SDXL 0.9 is the most advanced version of the Stable Diffusion series, which started with Stable Diffusion XL beta in April 2021. SDXL 0.9 boasts a 3.5B parameter base model and a 6.6B parameter model ensemble pipeline, which means it has more processing power and can create more detailed images than its predecessor.

SDXL 0.9 also leverages two CLIP models, including one of the largest OpenCLIP models trained to date (OpenCLIP ViT-G/14), which enhances its ability to create realistic imagery with greater depth and a higher resolution of 1024x1024.

How to use SDXL 0.9?

You can access SDXL 0.9 via ClipDrop by Stability AI, a web app that allows you to create stunning visuals with just a few clicks. ClipDrop also offers other tools to enhance your images, such as cleanup, remove background, relight, image upscaler, replace background, text remover, reimagine XL, uncrop and more.

To use SDXL 0.9 on ClipDrop, you just need to enter a text prompt and click on the generate button. You can also add negative prompts to exclude unwanted features from your image.

Prompt: A sloth, hyperrealistic, sipping pina colada by the pool. UHD.

Prompt: A woman in a fashion magazine photoshoot sitting on a komodo dragon. fierce. calm. high fashion. hyperrealistic.

You can also use image-to-image prompting (inputting one image to get variations of that image), inpainting (reconstructing missing parts of an image), and outpainting (constructing a seamless extension of an existing image) with SDXL 0.9.

How to create images with SDXL 0.9 in NightCafe studio?

Another way to access SDXL 0.9 is via NightCafe studio, an AI art generator and community app that offers multiple methods of AI art generation. This is one of my favorite AI art generators. NightCafe studio has presets for SDXL 0.9 and SDXL Beta that allows you to choose different styles and options for your images. I have been using the SDXL Beta since it launched on NightCafe Studio.

To use SDXL 0.9 on NightCafe Studio, you just need to select the Stable Diffusion XL 0.9 option from the menu beside the SDXL beta and enter a text prompt. You can also choose from various presets made for SDXL such as Striking, Anime v2, Epic Origami, Hyperreal, Vibrant to customize your image.

What are some examples of SDXL 0.9?

Here are some examples of the prompts I tested on both SDXL beta (left) and 0.9 (right) in NightCafe Studio to show how far this model has come in just a few months.

Prompt: A woman with detailed purple futuristic dress. Masterpiece design dress. In a cool wavy swirl background with gradient. Uhd. Hyperdetailed. Insane quality. Select Hyperreal preset.

Prompt: A beautiful lady sitting on a crocodile in a high fashion magazine photo shoot, beautiful face and eyes. Uhd. Dramatic. Select NightCafe preset.

Prompt: A beautiful peacock, origami. Select Epic Origami preset.

I greatly appreciate how a simple and concise prompt can generate a remarkable image in SDXL 0.9. The Epic Origami preset is my preferred choice, and the peacock it creates is truly stunning.

How to get started with SDXL 0.9?

If you are interested in trying out SDXL 0.9 for yourself, you can sign up for ClipDrop or NightCafe Studio.

SDXL 0.9 is a revolutionary leap in AI-generated imagery that opens up new possibilities for creative expression and innovation. Whether you are a filmmaker, a musician, a designer, or just someone who loves to explore new things like me, you will find something amazing with SDXL 0.9.

What are you waiting for? Try it out today and share your creations with me by tagging me on Twitter @zeng_wt! I love to see what you can do with SDXL 0.9!

Follow me and subscribe to PicAisso newsletter for more AI art exploration and WhoWhatWhyAi weekly newsletter and WhoWhatWhyAi Daily for AI-related content.




I write about AI tools and sharing my workflow here. Expect some random AI fun and my AI build post!