Join our daily and weekly newsletters for the latest updates and exclusive content on our industry-leading AI coverage. He learns more
Creating generative AI images can be fun and useful, but that’s not all organizations need.
Creating text-to-image conversion for an organization isn’t just about creating images. It’s about integration with existing workflows and other enterprise AI tools. This is the trend that Stability AIthe vendor behind Stable Diffusion understands this.
Today, Stability AI and Amazon Web Services (AWS) jointly announced this Stable spread 3.5 large Available on Amazon Bedrock. AWS is the only public cloud service to offer leading Stability AI models.
This step isn’t just about simple availability. It’s about integration and a go-to-market strategy that breathes more life into Stability AI’s efforts as a company New CEO Brings a renewed focus on meeting customer needs. Amazon Bedrock provides a single, unified API that allows organizations to access and use multiple AI models, including Stable Diffusion. This is important, as AWS’s own research shows that most organizations use more than one model at the same time. It’s an approach that users, including the National Football League (NFL) and Stride Learning, are already benefiting from.
Amazon Bedrock’s deployment of Stable Diffusion 3.5 comes as Stable AI faces an increasingly competitive landscape with rivals including Google, Midjourney, Ideogram, Flux Pro from Black Forest Laboratoriesamong other things. The company aims to stand out with greater image diversity in terms of style, agile commitment and enterprise workflow.
“There’s a reason we’re at AWS,” Prem Akkaraju, CEO of Stability AI, told VentureBeat in an exclusive interview. “That’s because this is where developers and creators are, and we want to bring our tools and models where they are. Our goal is to empower professional content creators.
AWS has its own image generation technology, why do you need Stable Diffusion?
The increasingly competitive text-to-image AI landscape also includes models from Amazon. At the beginning of this December Amazon Nova AI A family of models has been announced, including image generation models.
Bhaskar Sreedharan, vice president of AI, ML and Infrastructure Services at AWS, told VentureBeat that having multiple text-to-image models provides user choice. Amazon Bedrock provides a single, unified API for users, so they can choose to publish any model available on the platform using the same API. Sridharan also noted that AWS provides sample evaluation tools that can help organizations choose the best tool for a specific deployment.
Not surprisingly, Akkaraju finds the Stable Diffusion 3.5 outperforming other models. This is confirmation that AI for stability has been supported by the reported criteria for immediate commitment.
“We have shown in our research that Stable Diffusion 3.5 Large leads the market in instant commit, allowing models to closely follow a specific text prompt and truly making it the best choice for efficiency and high-quality performance,” said Akaraju.
How organizations can integrate Stable Diffusion into their Amazon Bedrock AI workflow
Stable Diffusion 3.5 has been made available to Stability AI users via the company’s API as well as the Stable Assistant service since late October.
Stability AI sees no real difference in price between using Stable Diffusion on its API or via Amazon Bedrock. The real benefit for enterprise users with Stable Diffusion now available on Amazon Bedrock is the ability to fit into larger, more complex enterprise workflows. Organizations can benefit from a unified workflow that connects multiple models from different vendors together 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 to create custom cleat designs. The app uses both Anthropic Claude and Stable Diffusion, Sreedharan explained. The NFL uses Cloud to create detailed prompts that understand the user’s preferences and determine what they want. This prompt is then entered into Stable Diffusion to generate the image. The entire process and workflow takes place on Amazon Bedrock without having to move between different services.
Another organization that has benefited from the integration is Education Vendor Learning step. The company needed images to support its online educational game Legends Library — lots of images, up to 1,000 per minute. Amazon Bedrock provides the infrastructure to support this scale to run Stable Diffusion. Beyond just high performance, there was also a need to secure image generation output. This is where Amazon Bedrock Guardrails API It fits. By using guardrails, Stride Learning is able to meet responsible AI policies for image generation, Sreedharan noted.
“When you do all this with a single API endpoint, it makes it very easy for customers to build these types of applications,” Sreedharan said.
AI stability for an eventful 2024 and the road ahead for its new leadership
The Stable Diffusion 3.5 update and availability on Amazon Bedrock caps an eventful year for Stability AI.
Company founder and former CEO Imad Mostafa resigned in March amid concerns about concentration and lack of revenue. It took until June for Stability AI to find a permanent replacement by hiring Akkaraju.
Up until now, Akkaraju has been responsible for a series of model updates. It has also helped attract new investors such as Napster founder Sean Parker and advisors including the famous director James Cameron In September. Akaraju comes from a visual effects background and has helped make films including Cameron’s Avatar. In his view, the professional visual media industry will completely transform over the next few years, moving to created content rather than displayed content.
“It’s a good thing we work a lot in the creative business, because these are probably some of the most demanding clients we could ask for,” Akaraju said.
Looking to the future, he joked that Stability AI’s plan is to take over the world. On a more serious note, he expects innovation to continue as his company strives to meet real workflow needs.
“We will continue to push the model forward,” Akaraju said. “You may see us do another release next year for the next image model and stay on the crest of the wave.”
https://venturebeat.com/wp-content/uploads/2024/12/stable-diffusion-inbedrock-smk.jpg?w=1024?w=1200&strip=all
Source link