Snowflake has announced dynamic model routing within Cortex AI Gateway and its flagship AI products, and expanded its catalog with leading open models. The innovations help companies cut unnecessary AI spend and improve what it calls intelligence efficiency, a metric that measures how effectively organizations translate compute capacity, AI models, their data, and business context into real business value.
The new capabilities build on Cortex AI Gateway, which Snowflake announced in July as a unified foundation to govern agent connections, route requests intelligently, and optimize AI consumption. As companies deploy more AI applications and agents in production, using the same model for all tasks can drive costs, while evaluating and managing an increasingly diverse mix of models creates more work for development teams. Snowflake addresses both challenges by automating model selection and giving customers access to a broader range of models, both open-source and proprietary.
AI Agent Capabilities
With the addition of dynamic model routing, Cortex AI Gateway can automatically select the specific model that offers the optimal balance between quality and cost for the given task. Dynamic model routing is also integrated into Snowflake’s flagship AI products, including Snowflake CoCo and Snowflake CoWork, and is available to third-party AI agents that use Cortex AI Gateway.
This new functionality routes repetitive or lower-complexity tasks to more efficient models, while tasks requiring higher reasoning capacity are routed to frontier models. This helps customers reduce unnecessary inference spend without having to manage model selection for each request themselves. Snowflake will also extend customer access to leading open models, such as DeepSeek-V4-Flash 07311 and GLM-5.33, through Snowflake Cortex AI. This adds to Snowflake’s broad model library, giving customers even more options to balance model quality and cost while keeping governed data secure within Snowflake.
Snowflake’s latest innovations give companies greater control over AI cost efficiency as they scale usage, helping them boost intelligence efficiency by assigning each task to the right model and trimming unnecessary spending.
“Companies are increasingly rigorous with respect to the costs and ROI of AI. The question isn’t how much AI they’re using anymore, but whether that AI translates into meaningful business value,” says Sridhar Ramaswamy, Snowflake’s CEO. “Achieving intelligence efficiency requires the flexibility to use the best model for each task as the landscape evolves. Snowflake’s role is to absorb that complexity so customers can focus on outcomes while we optimize the model selection in the background.”
“Companies are overwhelmed by so many model options, but the real issue isn’t which model to pick. It’s the operational burden of selecting the right one for each task, at scale. Snowflake’s dynamic model routing directly addresses that gap,” notes Sanjeev Mohan, director and founder of SanjMo. “By automating intelligent model selection within Cortex AI Gateway, Snowflake removes a real friction point that has been slowing enterprise AI deployments. The ability to tailor workload complexity to the cost of the model, without rebuilding infrastructure every time a new model appears, is exactly the kind of efficiency companies need to move from experimentation to AI at scale.”
