Snowflake and Anthropic partner up on agentic AI, model sharing
The pair will bring Anthropic’s Claude models to customers via Snowflake’s AI platform
Snowflake and Anthropic have signed off on a multi-year partnership that will deliver AI models to customers via the Snowflake platform, Snowflake has announced.
Anthropic’s Claude 3.5 models will be available on Snowflake’s Cortex AI, a managed service that allows Snowflake users to leverage large language models (LLMs) from other developers. Snowflake’s agentic AI products — Snowflake Intelligence and Cortex Analyst — will also leverage Claude models under the new partnership.
Snowflake said that combining the capabilities of Claude with Snowflake’s platform will allow businesses to use their data more effectively and develop conversational assistance and large-scale language processing.
Claude’s LLMs will allow data agents in Snowflake to analyze data, run ad-hoc data analytics, generate visualizations, and deploy multi-step workflows, Snowflake said.
As Snowflake Horizon Catalog’s compliance, security, and privacy capabilities are built into Cortex AI, users will also be able to deploy Claude with enhanced access controls and governance guardrails.
AI access for all
Michael Gerstenhaber, vice president of product at Anthropic, said the partnership would allow businesses of any size or industry to have access to the most advanced models.
"Together, we're helping Snowflake's customers to build transformative applications that use Claude's superior reasoning and problem-solving capabilities to tackle their most complex business challenges," he added.
Building an Anthropic model into Cortex AI means users will more easily be able to "talk" to their enterprise data, Snowflake said.
RELATED WHITEPAPER
Users will be able to develop AI applications that deliver reliable responses and avoid hallucinations using an organization's entire knowledge base, such as customer documentation or technical specifications.
Snowflake’s enterprise AI products will be optimized for Claude "out-of-the-box," the firm said, and Snowflake staff will also be using Claude internally to create agentic workflows.
-
Why the US imposed export controls on Anthropic’s Fable and Mythos models – and why they’ve been liftedNews Anthropic tightens up safeguards and offers expanded early access to US government to end export control issues
-
Anthropic touts new Claude Sonnet 5 model range, offering performance ‘close to that of Opus 4.8, but at lower prices’ – here’s what users can expectNews Claude Sonnet 5 comes with intuitive agentic capabilities, performance boosts, and cost-efficient ‘effort levels’
-
Using data to help deal with ever-changing and unpredictable weather conditionsCase study Ordnance Survey and Snowflake have partnered on the creation of an IFRM to help better identify flood risks
-
Copilot Cowork is now generally available: Everything you need to know, including pricing, usage limits, and new featuresNews A host of partner plugins are already available for Copilot Cowork, and more are coming
-
Anthropic suspends Fable and Mythos systems "for all users" after US government claims jailbreak riskNews Despite complying with the government, Anthropic suggests it's only a "potential narrow, non-universal jailbreak"
-
Anthropic just launched Claude Fable 5, its first Mythos-class AI model – but it has new safeguards to prevent misuse and will ‘fall back’ to Opus 4.8 for ‘high risk’ queriesNews The launch of Claude Fable 5 marks the first public release of a Mythos-class AI model
-
‘We experimented with efforts to differentially reduce these capabilities’: Anthropic toned down Opus 4.7’s cyber uses in wake of Claude Mythos releaseNews Testing of new cyber-related safeguards for Anthropic’s Opus 4.7 model could shape the future public release of Claude Mythos
-
Anthropic is worried hackers could abuse its Claude Mythos AI model – so it's asking big tech partners to test it behind closed doorsNews Anthropic’s Project Glasswing will give a host of leading tech companies access to its new Claude Mythos model for testing
