Snowflake announced a $6 billion agreement with Amazon Web Services (AWS) to increase the use of AWS’s ARM-based Graviton CPU chips and GPU-accelerated instances over five years starting in 2026 [1, 2, 3, 4]. The deal marks Snowflake’s largest cloud spending commitment to date and is close to its total prior sales through AWS Marketplace of $7 billion since 2012 [1, 4].
Snowflake plans to use AWS Graviton processors primarily for general purpose compute tasks and Nvidia GPUs for AI model training and inference [1, 2, 4]. Amazon CEO Andy Jassy said the company’s own AI chips provide "better price-performance" than Nvidia’s offerings [1]. Snowflake has been shifting toward AI with its Cortex AI suite, which includes natural language query capabilities and AI coding agents [1, 4].
Snowflake experienced a surge in cloud usage after its customers doubled AWS spending to $2 billion in 2025, a major driver of Snowflake’s AWS services growth [1]. The enterprise reported strong fiscal Q1 results for the period ended April 30, with earnings per share of $0.39 and revenue of $1.39 billion, beating analyst estimates [2].
Snowflake will also expand its AWS presence by adding 10 new regions worldwide, enhancing its infrastructure footprint [4]. Snowflake first adopted AWS Graviton processors in 2022 and announced a partnership with Nvidia for AI acceleration in 2023 [2]. AWS introduced its ARM-based Graviton chip in 2018, which is the company’s most successful custom silicon to date [2].
Snowflake CEO Sridhar Ramaswamy said, "We are moving into the era of the agentic enterprise, where AI systems don’t just answer questions, but help organizations reason over trusted data, coordinate workflows, and drive real business outcomes. With AWS, we are making it easier for enterprises to bring AI directly to governed data" [4].
