AI data cloud giant Snowflake has extended its partnership with AI developer Anthropic in a $200MN deal that brings Claude-powered AI agents deeper into Snowflake Cortex AI, helping enterprises analyze and act on customer data more effectively.
The expanded partnership will drop Claude models straight into the Snowflake platform for more than 12,600 customers across the three major clouds and launch a joint global go-to-market (GTM) initiative push aimed at scaling AI agents across some of the world’s biggest enterprises.
Claude Sonnet 4.5 is stepping in as the brains behind Snowflake’s enterprise intelligence agent, Snowflake Intelligence, to essentially provide a single, trusted AI that pulls together everything an enterprise knows and helps teams innovate more easily. Users can analyze, reason and extract insights from structured and unstructured data using simple natural language prompts, turning data into actionable knowledge.
Thousands of Snowflake customers are already processing trillions of Claude tokens per month through Snowflake Cortex AI, according to the companies.
This next stage is all about rolling out AI agents that can tackle multi-step, high-context work using Claude’s advanced reasoning and its ability to handle sensitive data securely. These agents can sift through financial, operational and customer information, make sense of the bigger picture, and show their work instead of just throwing back an answer. And because all of this happens inside Snowflake’s governed data environment, teams in tightly regulated industries like finance, healthcare, and life sciences can move from pilots to real production while ensuring compliance.
Scaling Claude for Enterprise With Snowflake
Anthropic has been pushing into the enterprise arena, especially when it comes to customer-facing AI. The company now leads the enterprise AI market, expanding from fewer than 1,000 business customers just two years ago to more than 300,000 worldwide, fueled by growing demand for its Claude models across industries.
The developer has been rolling out faster, more affordable models built for real-time interactions. These are the kind of tools that let large organizations run high-volume support, sales and service workflows without breaking the budget or sacrificing quality. It’s a sign that Anthropic is engineering its models for scale, reliability and the day-to-day demands of enterprise customer interactions. Dario Amodei, CEO and Co-Founder of Anthropic, said:
“Enterprises have spent years building secure, trusted data environments, and now they want AI that can work within those environments without compromise. This partnership brings Claude directly into Snowflake, where that data already lives. It’s a meaningful step toward making frontier AI genuinely useful for businesses.”
The partnership also unlocks deeper analysis through multimodal AI. With Snowflake Cortex AI Functions, organizations can tap into Anthropic’s latest models, including Claude Opus 4.5, which Snowflake hosted from its release. This allows users to query all kinds of data, from traditional rows and columns of numbers and text to images and audio, while leveraging SQL, the familiar language they already use to manage their data.
Snowflake Cortex Agents make it possible to build custom, production-ready multi-agent solutions. With Anthropic’s focus on safety, Claude models are designed to maintain high standards of accuracy and governance across workflows. Using Anthropic models within Snowflake, teams have access to Horizon Catalog, which provides end-to-end governance and responsible AI controls, so that they can move AI agents from pilot to production securely.
Snowflake also uses Claude Code internally to enhance developer productivity and accelerate innovation across its engineering organization. And Claude models power Snowflake’s internal GTM AI Assistant, which it built using Snowflake Intelligence, to enable sales teams to centralize data across the organization and ask questions in natural language to find relevant insights that help accelerate deal cycles.
