Amazon Web Services, which has been building its own AI training chips for years now, just introduced a new version known as Trainium3 that comes with some impressive specs.

The cloud provider, which made the announcement Tuesday at AWS re:Invent 2025, also teased the next product on its AI training product roadmap: Trainium4, which is already in the works and will be able to work with Nvidia’s chips.

AWS used its annual tech conference to formally launch Trainium3 UltraServer, a system powered by the company’s state-of-the art, 3 nanometer Trainium3 chip, as well as its homegrown networking tech. As you might expect, the third-generation chip and system offer big bumps in performance for AI training and inference over the second-generation chip, according to AWS.

AWS says the system is more than 4x faster, with 4x more memory, not just for training, but for delivering AI apps at peak demand. Additionally, thousands of UltraServers can be linked together to provide an app with up to 1 million Trainium3 chips — 10x the previous generation. Each UltraServer can host 144 chips, according to the company. 

Perhaps more importantly, AWS says the chips and systems are also 40% more energy efficient than the previous generation. While the world races to build bigger data centers powered by astronomical gigawatts of electricity, data center giant AWS is trying to make systems that drink less, not more.

It is, obviously, in AWS’s direct interests to do so. But in its classic, Amazon cost-conscious way, it promises that these systems save its AI cloud customers money, too.  

AWS customers like Anthropic (of which Amazon is also an investor), Japan’s LLM Karakuri, SplashMusic, and Decart have already been using the third-gen chip and system and significantly cut their inference costs, Amazon said. 

AWS also presented a bit of a roadmap for the next chip, Trainium4, which is already in development. AWS promised the chip will provide another big step up in performance and support Nvidia’s NVLink Fusion high-speed chip interconnect technology.  

This means the AWS Trainium4-powered systems will be able to interoperate and extend their performance with Nvidia GPUs while still using Amazon’s homegrown, lower-cost server rack technology.  

It’s worth noting, too, that Nvidia’s CUDA (Compute Unified Device Architecture) has become the de facto standard that all the major AI apps are built to support. The Trainium4-powered systems may make it easier to woo big AI apps built with Nvidia GPUs in mind to Amazon’s cloud.

Amazon did not announce a timeline for Trainium4. If the company follows previous rollout timelines, we’ll likely hear more about Trainium4 at next year’s conference.

Follow along with all of TechCrunch’s coverage of the annual enterprise tech event here.