Snowflake bets $6B on AWS to power next wave of agentic AI

Snowflake has signed an expanded multi-year strategic collaboration agreement with Amazon Web Services that includes a commitment to spend $6 billion on AWS infrastructure over the next five years. The spending will center on Amazon’s custom Graviton processors and GPU-based services to run Snowflake’s data cloud and AI workloads, including its Cortex AI platform for tasks such as text-to-SQL, summarization, and other generative and agentic AI use cases. This is Snowflake’s largest AWS commitment to date, up from earlier agreements of $1.2 billion at the time of its 2020 IPO and $2.5 billion in 2023, reflecting accelerating AI-driven demand on AWS.

The partnership goes beyond raw infrastructure consumption to include deeper technical and commercial integration between the two companies. Snowflake and AWS plan tighter product integrations across generative and agentic AI, giving joint customers more direct ways to run AI on governed enterprise data without moving it across systems. The agreement also expands co-selling and marketplace initiatives, building on more than $7 billion in lifetime Snowflake sales via AWS Marketplace and over $2 billion in Snowflake customer spend through AWS in calendar 2025 alone. In addition, Snowflake will broaden its footprint in AWS data centers, with launches completed or underway in 10 new regions including Auckland, Cape Town, Bangkok, and the AWS European Sovereign Cloud to support global and regulated workloads.

Strategically, the deal positions Snowflake as a flagship independent data and AI platform that is leaning into AWS’s silicon roadmap rather than building its own chips, while giving AWS a large, growing, and AI-intensive anchor tenant for Graviton and GPU capacity. The companies state that the joint goal is to help enterprises move from AI experimentation to production-scale, agentic AI systems that can reason over trusted data and automate workflows with strong governance. For Snowflake, the agreement locks in preferential access to AI infrastructure and reinforces its AI Data Cloud strategy at a time when it is also acquiring Natoma to extend governance and identity controls from data to AI agents and tools across the enterprise. For AWS, the commitment adds to a growing roster of large AI infrastructure deals and underscores Graviton’s role as a core CPU platform for data-heavy and AI-adjacent workloads.