If you are interested in generative AI and want to create stunning images from text prompts, you might have heard of Stability AI, a community-driven, open AI company that offers foundation models that can generate text, images, audio, video, code, and more from simple text instructions.
In this article, we will show you how to access and use Stability AI 1.0 on AWS Bedrock via three different methods: API, Amazon SageMaker JumpStart, and Stability AI’s apps (ClipDrop and DreamStudio).
What is Stability AI 1.0 on AWS Bedrock
Stability AI is known for its flagship text-to-image suite of models, Stable Diffusion, which can produce realistic and high-quality images from natural language descriptions.
You might also be familiar with AWS Bedrock, a fully managed service from Amazon Web Services (AWS) that provides access to foundation models from leading AI startups and Amazon Titan via an API.
AWS Bedrock enables customers to build and scale generative AI applications easier and faster, without having to worry about the infrastructure, security, and compliance issues.
How to Use Stable Diffusion XL 1.0 on AWS Bedrock via Amazon SageMaker JumpStart
Amazon SageMaker JumpStart allows you to choose Stability AI as a model provider and select Stability AI 1.0 as a foundation model.
To use Amazon SageMaker JumpStart, you need to launch it and configure the settings. You can do this by following the steps below:
Step 1: Go to Amazon SageMaker JumpStart website and click on “Get started with Amazon SageMaker JumpStart”.
Step 2: Log in with your AWS account or create a new one if you don’t have one already.
Step 3: Choose your preferred region and click on “Create notebook instance”.
Step 4: Wait for the notebook instance to be ready and then click on “Open Jupyter”.
Step 5: Click on “JumpStart” in the left sidebar and then click on “Models”.
Step 6: Choose Stability AI 1.0 as a model provider and then click on Stable Diffusion XL 1.0 as a foundation model.
Step 7: Click on “Create notebook” and then choose a name and a kernel for your notebook.
After continuing to subscribe, the process will continue.
Once you have created your notebook, you can use the Jupyter notebook interface to generate images from text prompts. You can do this by following the steps below:
Run the first cell of the notebook to import the necessary libraries and modules.
Run the second cell of the notebook to initialize the Stability AI client with your AWS Bedrock API key.
Run the third cell of the notebook to define a function that generates an image from a text prompt and displays it on the screen.
Run the fourth cell of the notebook to test the function with an example prompt.
Modify the prompt in the fourth cell or create new cells with different prompts to generate more images.
To save an image, right-click on it and choose “Save Image As…” from the menu. Choose a name and a location for your image file and click on “Save”.
You can also check out our blog, How to Use Stable Diffusion XL 0.9 in Clipdrop to Create Stunning Images for more tips and tutorials on How to Use Stable Diffusion XL 0.9 in Clipdrop.
How to Use Stable Diffusion XL 1.0
To use Stability AI’s apps, you need to download and install them on your device. You can do this by following the steps below:
- Go to clipdrop.co or dreamstudio.ai website and click on “Download” or “Get Started”.
- Choose your preferred platform such as iOS, Android, Windows, or Mac and follow the instructions to download and install the app.
- Launch the app and sign up with your email or social media account.
- Go to the settings and connect your app to AWS Bedrock by entering your API key.
Once you have connected your app to AWS Bedrock, you can use the app’s features such as inpainting, outpainting, image-to-image, and fine-tuning. You can do this by following the steps below:
To use inpainting, select an image from your gallery or capture one from your camera or screen. Use the eraser tool to erase some parts of the image that you want to fill in.
To use outpainting, select an image from your gallery or capture one from your camera or screen. Use the crop tool to crop some parts of the image that you want to extend.
To use image-to-image, select an image from your gallery or capture one from your camera or screen. Add some text prompts to modify the image such as changing the color, style, or content. Tap on the fine-tune button and wait for SDXL 1.0 to learn from your images and prompts.
To export an image, tap on the export button and choose a format such as PNG, JPG, or SVG. Choose a name and a location for your image file and tap on “Save” and share the image with friends and family.
FAQs
Conclusion
In this article, we have shown you how to access and use Stability AI 1.0 on AWS Bedrock via three different methods: API, Amazon SageMaker JumpStart, and Stability AI’s apps (ClipDrop and DreamStudio).
We hope that this article has inspired you to try out Stability AI 1.0 on AWS Bedrock and explore its possibilities for your own projects and applications.
Stability AI 1.0 on AWS Bedrock: A Step-by-Step Guide