Stable Diffusion 3.5 Large is now available in Amazon Bedrock | Amazon Web Services

As we have overtaken in AWS Re: Invent 2024, in Amazon in Amazon in the Amazon pad at night to create high -quality images of text descriptions in a wide range of styles to accelerate the creation of art, visual effects and images for customers in media , games, advertising and retail retail.

In October 2024, AI introduced a stability of a 3.5 large, large, most powerful model in a stable diffusion family with 8.1 billion parameters trained on Amazon Sagemaker Hyperpod, with superior quality and immediate adherence. Stable 3.5 large diffusion can speed up storyboarding, art concept and rapid prototyping of visual effects. You can quickly generate high quality 1-megapixel images for campaigns, social media and advertising posts, save time and resources while maintaining creative control.

Stable Diffusion 3.5 Large offers users of almost endless creative options, including:

• Versatile styles -You can generate pictures in a wide range of styles and aesthetics, including three -dimensional, photography, painting, line art and virtually any visual style you can imagine.
• Quick adhesion -Corched adhesion for quick adhesion can use a stable 3.5 large diffusion to carefully monitor its text challenge, making it the best choice for efficient and quality performance.
• Different outputs – You can create pictures that surround you with a diverse world, to introduce people with different tones and skin functions without the need for extensive challenges.

Today, the stable Ultra image in Amazon Bedrock has been updated to include a stable 3.5 diffusion in the underlying architecture of the model. Stable ultra image, powered by the most advanced AI stability models, including stable diffusion 3.5, sets a new standard when generating image. It stands out in typography, complex compositions, dynamic lighting, vibrant colors and artistic cohetisse.

With the latest update of stable diffusion models in Amazon Bedrock, you have a set of solutions to increase your creativity and speed up the workflows of generating images.

Start with a stable 3.5 -large diffusion in Amazon Bedrock
Before you start if you are a newcomer to use AI Stability models, go to the Amazon Bedrock Console and choose Access to the model On the lower left pane. You want to get access to the latest AI Stability models Stable diffusion of 3.5 large In stability AI.

If you want to test the AI ​​stability models in Amazon, choose Picture under Playground On the left pane. Then he chooses Select the model and choose Stability ai as a category of a Stable diffusion of 3.5 large as a model.

You can generate a picture with a challenge. Here is a challenge to create a picture:

High-energy street scene in a neon-lit Tokyo alley at night, where steam rises from food carts, and colorful neon signs illuminate the rain-slicked pavement.

By selecting View the API requestYou can also access examples of using the model code in the AWS (AWS CLI) and AWS SDKS command line. You can use stability.sd3-5-large-v1:0 Like a model ID.

If you want to get a picture with a single command, write the JSON output file to a standard output, and use JQ to extract the coded image to decodate at the run. The output is written in the img.png file.

Here is a sample of the AWS clinic:

$ aws bedrock-runtime invoke-model \
   --model-id stability.sd3-5-large-v1:0 \
   --body "{\"prompt\":\"High-energy street scene in a neon-lit Tokyo alley at night, where steam rises from food carts, and colorful neon signs illuminate the rain-slicked pavement.\",\"mode\":\"text-to-image\",\"aspect_ratio\":\"1:1\",\"output_format\":\"jpeg\",\"seed\":0}" \
   --cli-binary-format raw-in-base64-out \
   --region us-west-2 \
/dev/stdout | jq -r '.images(0)' | base64 --decode > img.jpg

Here’s how to use a stable Ultra 1.1 image to include a stable 3.5 diffusion in the background architecture of the SWK SDK for Python (Boto3). This simple interactive application Interactive asks to prompt to image text and then call Amazon Bedrock to generate an image stability.stable-image-ultra-v1:1 Like a model ID.

import base64
import boto3
import json
import os

MODEL_ID = "stability.stable-image-ultra-v1:1"

bedrock_runtime = boto3.client("bedrock-runtime", region_name="us-west-2")

print("Enter a prompt for the text-to-image model:")
prompt = input()

body = {
    "prompt": prompt,
    "mode": "text-to-image"
}
response = bedrock_runtime.invoke_model(modelId=MODEL_ID, body=json.dumps(body))

model_response = json.loads(response("body").read())

base64_image_data = model_response("images")(0)

i, output_dir = 1, "output"
if not os.path.exists(output_dir):
    os.makedirs(output_dir)
while os.path.exists(os.path.join(output_dir, f"img_{i}.png")):
    i += 1

image_data = base64.b64decode(base64_image_data)

image_path = os.path.join(output_dir, f"img_{i}.png")
with open(image_path, "wb") as file:
    file.write(image_data)

print(f"The generated image has been saved to {image_path}")

The application writes the resulting image in output The directory that is created if it is not present. In order to not exceed existing files, the code checks existing files to find the first name available with img_<number>.png format.

If you want to learn more, visit the Invoke API Example using AWS SDK to create applications and generate a picture using different programming languages.

Interesting example
Here are several pictures created with stable diffusion 3.5 wide.

Prompt: Full-body university students working on a tech project with the words Stable Diffusion 3.5 in Amazon Bedrock, cheerful cursive typography font in the foreground.	Prompt: Photo of three potions: the first potion is blue with the label "MANA", the second potion is red with the label "HEALTH", the third potion is green with the label "POISON". Old apothecary.
Prompt: Photography, pink rose flowers in the twilight, glowing, tile houses in the background.	Prompt: 3D animation scene of an adventurer traveling the world with his pet dog.

Now available
Stable 3.5 large model today is generally available in Amazon Bedrock in the USA in the West (Oregon) AWS. For future updates, see the entire Region list. If you want to learn more, check out the Amazon Bedrock page and on Amazon Bedrock Pricking on Amazon Bedrock.

Try a stable 3.5 diffusion in the Amazon Bedrock Today console and send feedback to AWS RE: Post for Amazon Bedrock or through the usual AWS support contacts.

– Channels

Updated 19 December 2024 – Fixed AWS CLA command to call up the model.