AWS Unveils Graviton5, Nova Models, and Autonomous Agents at re:Invent 2025
AWS announces a suite of major AI and chip innovations at re:Invent 2025, headlined by the powerful new Graviton5 processor and an expanded Amazon Nova model family. The event also introduces frontier agents capable of working autonomously for days to handle complex software development tasks.
AWS unveils a massive lineup of technological innovations at re:Invent 2025, led by the introduction of Graviton5, the company's most powerful and efficient custom CPU to date. The new Graviton5-based Amazon EC2 M9g instances deliver up to 25% higher performance than the previous generation, featuring 192 cores per chip and a cache that is five times larger. This release underscores AWS's ongoing commitment to custom silicon, as Graviton processors power more than half of all new CPU capacity added to the cloud platform for the third consecutive year.
The tech giant also significantly expands its Amazon Nova portfolio with four new models designed for reasoning, multimodal processing, conversational AI, code generation, and agentic tasks. A standout addition is Nova Forge, which pioneers an "open training" approach by giving organizations direct access to pre-trained model checkpoints to blend with their proprietary data. Furthermore, the new Nova Act achieves a breakthrough 90% reliability rate for browser-based UI automation, enabling companies like Reddit and Hertz to streamline operations and accelerate development velocity by up to five times.
In addition to hardware and foundation models, AWS introduces a new class of "frontier agents" that function as autonomous extensions of software development teams. These advanced agents, including the newly announced Kiro autonomous agent, represent a major step change in AI capabilities because they work continuously for hours or days without requiring human intervention. Rounding out the re:Invent announcements are the availability of Trainium3 UltraServers and AWS AI Factories, which allow customers to deploy advanced AI infrastructure directly within their own existing data centers.