Stable Diffusion 3.5 comes to Amazon Bedrock: What it means for enterprise AI workflows

Creating sophisticated generative AI images can be fun and useful, but that’s not all businesses need.

Enterprise text-to-image generation is more than just creating images. This is integration with existing workflows and other enterprise AI tools. That is a direction that AI Stabilitythe vendor behind Stable Diffusion, gets it.

Today, Stability AI and Amazon Web Services (AWS) jointly announced that Stable Diffusion 3.5 Large is available on the Amazon Bedrock service. AWS is the only public cloud service to offer flagship Stability AI models.

The measure is not just about simple availability. It’s about integration and a go-to-market strategy that breathes more life into Stability AI’s efforts as the company’s new CEO brings a renewed focus to meeting customer needs. Amazon Bedrock provides a single, unified API that allows businesses to access and use multiple AI models, including Stable Diffusion. This is important, as AWS’s own research shows that most companies use more than one model at a time. It’s an approach that users like the National Football League (NFL) and Stride Learning are already benefiting from.

Amazon Bedrock’s implementation of Stable Diffusion 3.5 comes as Stability AI faces an increasingly competitive landscape with rivals such as Google, Midjourney, Ideogram, and Black Forest Labs’ Flux Pro, among others. The company aims to differentiate itself with greater diversity of images in terms of style, rapid adherence and business workflows.

“There’s a reason we’re on AWS,” Prem Akkaraju, CEO of Stability AI, told VentureBeat in an exclusive interview. “It’s because that’s where the developers and creators are, and we want to bring our tools and our models to where they are. Our goal is to empower professional content creators.”

AWS has its own imaging technology, why do you need Stable Diffusion?

The increasingly competitive text-to-image generation AI landscape also includes models from Amazon. The Amazon Nova AI family of models, including imaging models, was announced in early December.

Baskar Sridharan, vice president of AI and ML infrastructure and services at AWS, told VentureBeat that having multiple text-to-image generation models gives the user options. Amazon Bedrock provides a single unified API for users, so they can choose to deploy any model that is available on the platform using the same API. Sridharan also noted that AWS provides model evaluation tools that can help companies choose the best tool for a specific implementation.

Not surprisingly, Akkaraju considers the Stable Diffusion 3.5 to be superior to other models. That’s a claim Stability AI has backed up through informed benchmarks of rapid compliance.

“In our research we showed that Stable Diffusion 3.5 Large leads the market in fast adhesion, allowing models to closely follow a given text prompt and truly making it the best choice for efficiency and high-quality performance,” Akkaraju said.

How companies can integrate Stable Diffusion into an Amazon Bedrock AI workflow

Stable Diffusion 3.5 large has been available to Stability AI users via the company’s API as well as its Stable Assistant service since late October.

Stability AI sees no real price difference between using Stable Diffusion in their API or through Amazon Bedrock. The real benefit for business users having Stable Diffusion now available on Amazon Bedrock is the ability to adapt to larger, more complex enterprise workflows. Enterprises can benefit from a unified workflow that links multiple models from different vendors with a single API.

It’s an approach the NFL is already using.

The NFL has an app called “My Cause, My Cleats” that uses Amazon Bedrock to enable a collaborative, community-focused experience for creating custom cleat designs. Sridharan explained that the app uses both Anthropic Claude and Stable Diffusion. The NFL uses Claude to create turn-by-turn prompts that understand user preferences and determine what they want. That message is then fed into Stable Diffusion to generate the image. The entire process and workflow is done in Amazon Bedrock without the need to jump between different services.

Another organization that has benefited from integration is the education provider. Stride learning. The company needed images to support its online learning game Legends Library: lots of images, up to 1,000 images per minute. Amazon Bedrock provides the infrastructure necessary to support that scale to run Stable Diffusion. Beyond high-performance scaling, there was also a need to ensure imaging output. That’s where the Amazon Bedrock Guardrails API fits in. Sridharan noted that with guardrails, Stride Learning can comply with responsible AI policies for image generation.

“When you do all this using a single API endpoint, it makes it very easy for customers to build these types of applications,” Sridharan said.

Stability AI’s eventful 2024 and the road ahead with its new leadership

The update and availability of Stable Diffusion 3.5 on Amazon Bedrock caps an eventful year for Stability AI.

The company’s founder and former CEO, Emad Mostaque, resigned in March amid concerns about focus and lack of revenue. It took until June for Stability AI to find a permanent replacement with the appointment of Akkaraju.

Until now, Akkaraju had been at the helm of a series of model updates. It has also helped attract new investors such as Napster founder Sean Parker and advisors, including acclaimed director James Cameron in September. Akkaraju has a background in visual effects and helped make films, including Cameron’s. Avatar. In his opinion, the professional visual media industry will completely transform in the coming years, moving to generated content instead of rendered content.

“It’s a good thing we work a lot in the creative business, because they are probably some of the most demanding clients we could ask for,” Akkaraju said.

Looking ahead, he joked that Stability AI’s plan is to take over the world. Seriously, though, expect continued innovation as your business strives to meet real-world workflow needs.

“We’re going to continue to push the model,” Akkaraju said. “You may even see us make another launch next year of our next image model and stay on top of the wave.”