Luma AI Unveils Ray 2 Video Model with AWS Partnership for Enhanced AI Video Generation

At the AWS re:Invent conference in Las Vegas, Luma AI announced its new Ray 2 Video Model, an advanced tool designed to generate high-quality videos from text and image prompts in as little as 10 seconds. The upgraded model, which expands video lengths from five seconds to up to one minute, is aimed at general consumers, creative professionals, and developers.

Additionally, Luma AI revealed a strategic partnership with Amazon Web Services (AWS) to integrate its models into Amazon Bedrock, AWS’s platform for building and deploying AI applications. The collaboration positions AWS as Luma AI’s preferred compute partner and marks a significant step in making generative video technology widely accessible.

Luma AI’s Ray 2: Speed and Quality Redefined

Ray 2 is touted as the fastest and most efficient video generative model available. Leveraging multimodal transformer architecture, the model produces cinematic, realistic videos with smooth camera movements and dynamic storytelling elements. From sweeping landscapes to detailed close-ups, Ray 2’s capabilities include creating physically accurate interactions between characters, animals, and objects based on natural language prompts.

“Video has become an essential form of expression and learning, and we set out to offer a service that helps everyone—from creatives to professionals—become fluent in this new visual AI medium,” said Amit Jain, CEO and Co-founder of Luma AI.

The Ray 2 Video Model will be accessible through Luma AI’s Dream Machine service, which launched in June 2024 and now boasts over 25 million registered users.

Partnership with AWS: Expanding Reach and Capabilities

The partnership with AWS aims to democratize access to Luma AI’s video models through Amazon Bedrock, a fully managed service that allows developers to incorporate foundation models into their applications via a single API.

“Luma AI video models are already making it easier for creators around the world to bring their visions to life. With Amazon Bedrock, we aim to democratize access to these models, allowing creators to focus on the creative process—no arts background needed,” said Jon Jones, Vice President and Global Head of Startups at AWS.

AWS will provide infrastructure for training Luma AI’s models through Amazon SageMaker HyperPod and AWS AI chips, including AWS Trainium and AWS Inferentia, which offer high-performance training and inference at lower costs.

A Tool for Creatives and Developers

Ray 2 is engineered to cater to a broad audience, from casual users experimenting with visual storytelling to professionals seeking advanced tools for high-quality video production. The model’s intuitive design and AI-powered capabilities aim to lower the barrier to entry for video creation, empowering users to produce complex, visually stunning content with ease.

Luma AI’s Rapid Growth

Founded in 2021, Luma AI has quickly emerged as a leader in visual AI. The company’s Dream Machine platform, which integrates text-to-video and text-to-image capabilities, has achieved widespread adoption and attracted $80 million in funding from prominent investors, including Andreessen Horowitz and General Catalyst.

With its partnership with AWS and the launch of Ray 2, Luma AI is set to further solidify its position as a pioneer in generative AI, offering innovative tools that redefine how videos are created and shared.

The Ray 2 Video Model will be available soon through Luma AI’s Dream Machine platform and Amazon Bedrock.