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

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

Text-to-image generation for businesses is about more than just creating images. It’s about integration into existing workflows and other AI tools for companies. This is a direction that Stability AIthe provider behind Stable Diffusion understands.

Stability AI and Amazon Web Services (AWS) jointly announced this today Stable diffusion 3.5 large is available through the Amazon Bedrock service. AWS is the only public cloud service to offer the flagship resiliency AI models.

The move isn’t just about easy availability. It’s about integration and a go-to-market strategy that brings more life to Stability AI’s efforts as a company new CEO refocuses on meeting customer needs. Amazon Bedrock provides a single, unified API that enables companies to access and use multiple AI models, including Stable Diffusion. This is important because AWS’s own research shows that most companies use more than one model at a time. Users like the National Football League (NFL) and Stride Learning are already benefiting from this approach.

Amazon Bedrock’s deployment of Stable Diffusion 3.5 comes at a time when Stability AI faces an increasingly competitive environment with rivals such as Google, Midjourney, Ideogram and others Flux Pro by Black Forest Labsamong other things. The company’s goal is to differentiate itself from the competition by providing more image diversity in terms of style, quick compliance and company operations.

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

AWS has its own imaging technology. Why does it need stable diffusion?

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

Baskar Sridharan, vice president of AI and ML services and infrastructure at AWS, told VentureBeat that several text-to-image generation models provide user choice. Amazon Bedrock provides users with a single unified API, allowing them to deploy any model 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 particular deployment.

Unsurprisingly, Akkaraju sees Stable Diffusion 3.5 as superior to other models. This is a claim that Stability AI has backed up with reported benchmarks for immediate compliance.

“We show in our research that Stable Diffusion 3.5 Large is a market leader in prompt adherence, enabling models to accurately follow a given text prompt and truly making it the first choice for efficiency and high-quality performance,” said Akkaraju.

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

Stable Diffusion 3.5 Large has been available to Stability AI users since late October through the company’s API and Stable Assistant service.

Stability AI sees no real price difference between using Stable Diffusion on its API or through Amazon Bedrock. The real benefit for enterprise users of having Stable Diffusion now available on Amazon Bedrock is the ability to integrate into larger, more complex enterprise operations. Companies can benefit from a unified workflow that connects multiple models from different vendors with a single API.

It’s an approach the NFL already uses.

The NFL has an application called My Cause, My Cleats that uses Amazon Bedrock to provide a collaborative, community-focused experience in creating custom cleat designs. Sridharan explained that the application uses both Anthropic Claude and Stable Diffusion. The NFL uses Claude to create detailed prompts that understand user preferences and determine what they want. This prompt is then fed into Stable Diffusion to produce the image. The entire process and workflow takes place on Amazon Bedrock without the need to switch between different services.

Another organization that has benefited from the integration is the education provider Step 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 to support this scale of stable diffusion operations. Beyond pure high-performance scaling, there was also a need to secure imaging output. There is that Amazon Bedrock Guardrails API fits in. Sridharan noted that with guardrails, stride learning is able to adhere to responsible AI guidelines for image generation.

“Doing this all through a single API endpoint makes it easy for customers to build these types of applications,” Sridharan said.

The eventful year 2024 for Stability AI and the path ahead with new leadership

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

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

To date, Akkaraju has been responsible for a number of model updates. He has also helped bring in new investors such as Napster founder Sean Parker and advisors including high-profile directors James Cameron in September. Akkaraju has a background in visual effects and has been involved in the production of films such as Cameron’s Avatar. In his opinion, the professional visual media industry will completely change in the next few years and switch to generated content instead of rendered content.

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

Looking ahead, he joked that Stability AI’s plan is world domination. Seriously, though, he expects continued innovation as his company strives to meet real-world workflow needs.

“We will continue to push the model forward,” Akkaraju said. “Maybe next year you’ll even see us release another image model and just stay at the top of the wave.”