Amazon CEO: Over one million Trainium chips in production, annual revenue in the billions of dollars

Amazon is forging a new multi-billion dollar revenue path in the NVIDIA-dominated market with its self-developed AI chips. At the recent AWS re:Invent conference, Amazon unveiled for the first time the specific scale of its self-developed AI chip business. In a social media post, Amazon CEO Andy Jassy stated clearly that the second-generation AI training chip, Trainium2, “already has enormous appeal and is a multi-billion dollar annualized business, with more than one million chips currently in production.” This statement quantifies for the first time the commercial success of Amazon’s self-developed chip, directly addressing market concerns about its competitiveness in the AI infrastructure field. Jassy emphasized that the core reason customers choose Trainium is its “highly attractive cost-performance ratio compared to other GPU options,” which is the classic Amazon business model: providing proprietary technology at a lower price. He revealed that currently, more than 100,000 companies are using Trainium, and the chip supports the majority of usage on Amazon’s AI development platform, Bedrock. Bedrock is a key AI service offered by Amazon, allowing enterprise customers to choose and build applications across multiple AI models. The Trainium chip has become the main computing hardware for this platform, meaning Amazon has successfully achieved vertical integration from hardware to service within its own ecosystem. Key Customer Anthropic’s Strong Support Behind Trainium’s multi-billion dollar revenue, adoption by key customers played a decisive role. According to a report by technology media outlet CRN, AWS CEO Matt Garman revealed in an interview that AI startup Anthropic is one of the largest users of Trainium chips. Garman said: “We’re seeing Trainium2 gain tremendous traction, especially from our partner Anthropic.” He further pointed out that, in a project known as “Project Rainier,” Amazon has deployed more than 500,000 Trainium2 chips to help Anthropic build its next-generation Claude series AI models. Project Rainier is Amazon’s most extensive AI server cluster to date, designed to meet Anthropic’s rapidly growing compute demands. Amazon is a major investor in Anthropic. As part of the investment agreement, Anthropic uses AWS as its primary cloud service provider for model training. This strategic partnership not only brings a stable revenue source for Trainium chips, but also provides Amazon with a benchmark case to demonstrate the performance and scale of its chips. Launch of Next-Generation Trainium3 to Challenge NVIDIA’s Ecosystem While consolidating its existing market, Amazon is accelerating its catch-up through technological iterations. At the re:Invent conference, the company officially released the new generation AI chip, Trainium3. According to Jassy, the new chip has achieved a significant leap in performance: “Compared to Trainium2, Trainium3 will provide at least 4.4x computational performance, 4x energy efficiency, and nearly 4x memory bandwidth.” Despite the remarkable progress, challenging NVIDIA’s market position remains formidable. NVIDIA’s moat lies not only in its GPU hardware, but also in its proprietary CUDA software platform, which has become the de facto standard for AI development. According to TechCrunch, rewriting AI applications for non-CUDA chips is a complex and costly endeavor. Nevertheless, Amazon appears to be planning countermeasures. Reports indicate that its forthcoming Trainium4 chips will be designed to work alongside NVIDIA GPUs in the same system. Whether this move will ultimately weaken NVIDIA’s market share or further strengthen its dominance in the AWS cloud remains to be seen. Risk Warning and Disclaimer The market involves risks, and investment requires caution. This article does not constitute individual investment advice nor does it take into account the specific investment objectives, financial situation, or needs of any particular user. Users should consider whether any opinions, views, or conclusions in this publication are suitable for their particular circumstances. Any investments made based on this are at your own risk.