New Amazon Nova Models Ramp Up Generative AI Performance

Amazon Web Services (AWS) has unveiled Amazon Nova, a cutting-edge suite of foundation models (FMs) for generative AI. The Nova family is designed to offer unmatched performance and cost-efficiency, exclusively available on Amazon Bedrock.

Amazon Nova models cater to a wide spectrum of applications, explained Danilo Poccia, Chief Evangelist for AWS, in a blog post. "You can build on Amazon Nova to analyze complex documents and videos, understand charts and diagrams, generate engaging video content, and build sophisticated AI agents, from across a range of intelligence classes optimized for enterprise workloads," he wrote.

The Nova suite includes two primary categories: Understanding Models, which analyze text, images, and videos, and Creative Content Generation Models, designed for producing high-quality visuals and videos.

Understanding Models

  • Nova Micro: A lightweight text-only model focused on speed and cost, ideal for summarization, translation, and coding tasks.
  • Nova Lite: A multimodal model for real-time analysis of text, images, and videos, supporting customer interactions and document processing.
  • Nova Pro: A highly capable multimodal model excelling in complex workflows, including financial document analysis and code processing.
  • Nova Premier (coming in 2025): Expected to set new benchmarks for reasoning and customization.

Nova models are optimized for tasks such as Retrieval-Augmented Generation (RAG) and function calling. They are also customizable, allowing businesses to tailor the models to their specific needs, from brand voice alignment to specialized industry terminology.

Creative Content Generation Models

  • Nova Canvas: A state-of-the-art tool for generating studio-quality images with precision editing capabilities.
  • Nova Reel: A video generation model for creating professional-quality short films based on text prompts or reference images.

"These models bring creativity and precision to industries like marketing, entertainment, and ecommerce," Poccia said. "Nova Canvas and Nova Reel ensure enterprises can create compelling content at scale."

Amazon Nova models demonstrate superior performance on industry benchmarks, including TextVQA for visual question answering and TIFA for text-to-image evaluation. They support inputs up to 300,000 tokens and can analyze videos up to 30 minutes long.

The Nova models integrate seamlessly with Amazon Bedrock, enabling users to deploy and scale AI applications effortlessly. Features such as real-time streaming and batch processing allow businesses to adapt Nova to varied use cases, from customer service automation to marketing asset creation.

With robust safety features, Nova models prioritize responsible AI use. Content moderation, digital watermarking, and protections against misinformation ensure the technology is both secure and ethical. "We're building AI responsibly, with safeguards that match the capabilities of our models," Poccia emphasized.

Nova models support over 200 languages, including English, Spanish, Mandarin, and Hindi, allowing businesses to operate across geographies without needing separate AI systems for each region.

Amazon Nova's debut marks a significant milestone in AWS's efforts to dominate the generative AI landscape, Poccia said. By combining cutting-edge capabilities with enterprise-friendly pricing, AWS positions itself as a formidable challenger to industry giants like Microsoft and Google.

Poccia encouraged businesses to explore Nova through the Bedrock console. "This is just the beginning," he said. "We're excited to see how the global community leverages these tools to innovate and scale."

For more information, go to the AWS site.

About the Author

John K. Waters is the editor in chief of a number of Converge360.com sites, with a focus on high-end development, AI and future tech. He's been writing about cutting-edge technologies and culture of Silicon Valley for more than two decades, and he's written more than a dozen books. He also co-scripted the documentary film Silicon Valley: A 100 Year Renaissance, which aired on PBS.  He can be reached at [email protected].

Featured

  • split-screen digital illustration of two AI-influenced classrooms

    What AI Gets Right and How It Will Be Used in the Year Ahead

    AI has tremendous potential to do good in education while honoring and upholding the essential role of teachers. However, its success will depend on how we choose to use it.

  • DreamBox Math

    Discovery Education Announces Updates to Experience, DreamBox Math

    K-12 learning solution provider Discovery Education has announced enhancements to its Discovery Education Experience and DreamBox Math products, designed to create a more personalized, engaging learning experience for students.

  • The AI Show

    Register for Free to Attend the World's Greatest Show for All Things AI in EDU

    The AI Show @ ASU+GSV, held April 5–7, 2025, at the San Diego Convention Center, is a free event designed to help educators, students, and parents navigate AI's role in education. Featuring hands-on workshops, AI-powered networking, live demos from 125+ EdTech exhibitors, and keynote speakers like Colin Kaepernick and Stevie Van Zandt, the event offers practical insights into AI-driven teaching, learning, and career opportunities. Attendees will gain actionable strategies to integrate AI into classrooms while exploring innovations that promote equity, accessibility, and student success.

  • A geometric pattern of open Chromebook computers with bold outlines, subtle shading, and soft gradients, spaced evenly with vibrant green and blue accents on a neutral background.

    Challenges and Opportunities Ahead for the 'Great Chromebook Refresh'

    During the pandemic, the education community scrambled to provide students with laptops to promote online learning equity and mitigate learning loss. Today, those devices are approaching the end of their useful lives — and a "great Chromebook refresh" has been predicted as schools seek to replace them with newer models.