Snowflake Says Cortex AI Gateway Routing Delivers 3x Token Efficiency for AI Workloads

Snowflake said its AI model routing technology can deliver up to three times greater token efficiency than using a single advanced AI model for a workload, a claim the company tied to a test of its Cortex AI Gateway. Business Today reported the claim on October 26, 2026.

How Snowflake's Cortex AI Gateway Routing Works

The company said the test produced the efficiency gain while maintaining comparable output quality. The technology automatically directs tasks to suitable AI models based on their capabilities and the requirements of each workload, rather than relying on a single model for every request.

Snowflake announced dynamic model routing in August 2026 as businesses sought ways to make AI deployments more economical without compromising output quality. The capability addresses a challenge for businesses moving AI applications from pilot projects into production, where AI agents can make multiple model calls and interact with tools to complete a task.

"The economics of AI are now boardroom math," Baris Gultekin, Vice President of Applied AI at Snowflake, told Business Today. "As enterprises invest more in deploying AI across the business, they're starting to ask difficult questions about whether the value it creates justifies the cost," he said.

Gultekin said the goal should be to make AI useful in more places across the business, which means being disciplined about where additional intelligence makes a difference and where it only adds cost. He said businesses should start by defining what a task requires and then evaluate models against those requirements, citing quality, cost, latency, reliability, availability, security, governance and how well a model works with an enterprise's own data and context.

Snowflake's routing system is designed to restrict model choices to those approved by administrators and provide visibility into which model handles each request, while respecting existing data security and residency controls. Gultekin said giving an agent the right data, choosing an appropriate model for the task, and reducing unnecessary steps can all improve the outcome and its cost.

Snowflake's Anthropic Partnership and Funding Records

Snowflake deepened its strategic partnership with Anthropic at an estimated $200 million level in June 2026, making Claude Fable 5 natively available through Cortex AI, according to akta.pro company records. The same records list a $3.8 billion post-IPO debt round in September 2026 and a $376.1 million post-IPO equity round in February 2026.

Earlier Snowflake AI Product Launches

Snowflake introduced Decision Fast on October 19, 2026, a low-latency variant of its Decision models for real-time classification, scoring and routing, with REST API support via Cortex AI Gateway and decisions completing in under 300 ms, according to Snowflake. Snowflake put Horizon Context, a governed context layer in Horizon Catalog, into public preview on October 20, 2026, with metadata connectors for 18 sources, according to WebProNews. Snowflake launched the AIM application in public preview on October 20, a standalone desktop tool for data migration, according to Snowflake.

Snowflake's Next Steps

Snowflake's dynamic model routing is available through Cortex AI Gateway, which automatically selects the model suited to a task from models approved by the customer.