At the AWS re:Invent conference in Las Vegas, Amazon announced the launch of Amazon Nova, a new generation of multimodal generative AI models capable of producing text, images, and videos.
CEO Andy Jassy introduced six models under the Nova brand, including Amazon Nova Micro for text processing and Amazon Nova Reel for video generation. These models support 200 languages and are reported to be at least 75% more cost-effective than competitors.
In addition, Amazon plans to release a speech-to-speech model and a native multimodal-to-multimodal model by 2025. The company is also developing a large AI compute cluster utilizing its Trainium 2 chips, in partnership with Anthropic, expected to be the largest of its kind.
Amazon Bedrock, the platform for AI applications, received updates including Automated Reasoning Checks to minimize factual errors and Model Distillation for optimizing model size and performance. AWS highlighted its collaboration with clients like PwC and Moody’s, showcasing Bedrock’s ability to enhance business operations through advanced AI capabilities.