Snowflake (NYSE: SNOW), the AI Data Cloud company, today announced dynamic model routing2 within Cortex AI Gateway and Snowflake’s flagship AI products, alongside expanded access to leading open models. Snowflake’s innovations help enterprises reduce unnecessary AI spend and improve what Snowflake calls intelligence efficiency, measuring how effectively companies turn compute, models, data, and context into business impact.

This press release features multimedia. View the full release here: https://www.businesswire.com/news/home/20260818448905/en/

New capabilities in Cortex AI Gateway help enterprises lower AI costs with targeted model choice for each workload while keeping data governed in Snowflake

New capabilities in Cortex AI Gateway help enterprises lower AI costs with targeted model choice for each workload while keeping data governed in Snowflake

The new capabilities build on Cortex AI Gateway, which Snowflake announced in July 2026 as a unified foundation for governing agent connections, intelligently routing requests, and optimizing AI consumption. As companies deploy more AI apps and agents into production, using the same model for every task can drive up costs, while evaluating and managing a growing mix of models creates more work for development teams. Snowflake is addressing both challenges by making model selection more automatic and giving customers access to a broader range of open source and proprietary models.

With the addition of dynamic model routing, Cortex AI Gateway can automatically select the specific model with the optimal balance of quality and cost for the task at hand. Dynamic model routing is also integrated across Snowflake’s flagship AI products, including Snowflake CoCo and Snowflake CoWork, and is available to third-party AI agents using Cortex AI Gateway. This new capability directs lower-complexity or repetitive tasks to more efficient models, while work that requires deeper reasoning is routed to frontier models. This helps customers reduce unnecessary inference spend, without having to manage model selection for each request themselves. Snowflake will also be expanding customers’ access to leading open models, including DeepSeek-V4-Flash 07311and GLM-5.33, through Snowflake Cortex AI. This adds to Snowflake’s extensive 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 enterprises more control over the economics of AI as usage scales, helping them increase intelligence efficiency by matching each task with the right model and reducing unnecessary spend.

“Enterprises are becoming much more rigorous about the economics of AI. The question is no longer how much AI they are using, but whether that AI is translating into meaningful business value,” said Sridhar Ramaswamy, CEO, Snowflake. “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 model choice underneath.”

“Enterprises are drowning in model choices, but the real problem isn’t which model to pick. It’s the operational overhead of picking the right one for every task, at scale. Snowflake’s dynamic model routing directly addresses that gap,” said Sanjeev Mohan, Principal and Founder, SanjMo. “By automating intelligent model selection within Cortex AI Gateway, Snowflake is removing a real friction point that has been slowing enterprise AI deployment. The ability to match workload complexity to model cost, without rebuilding your infrastructure every time a new model drops, is exactly the kind of efficiency enterprises need to move from AI experimentation to AI at scale.”

Cortex AI Gateway Delivers Better AI Economics Through Model Choice

The addition of dynamic model routing in Cortex AI Gateway is designed to help enterprises use costly frontier models only when they materially improve an outcome across Snowflake CoCo, Snowflake CoWork, and third-party AI agents. Beyond automatically selecting a model for each request, dynamic model routing gives customers control over which models and providers are available to their users. This is particularly important for global organizations navigating regional model availability or for regulated industries with specific compliance requirements. As model performance and pricing change, Cortex AI Gateway can update routing decisions across Snowflake’s AI products like Snowflake CoCo and Snowflake CoWork so customers don’t need to rebuild their apps or agents. This allows enterprises to take advantage of new model options as they emerge while Snowflake manages the complexity of model choice and optimization underneath.

Snowflake’s internal testing indicates that using a mix of open and proprietary models for various tasks can deliver comparable quality, while materially improving token efficiency. In one evaluation, agents using dynamic model routing with Cortex AI Gateway built a dbt pipeline with up to 3x greater token efficiency than a frontier-model-only path, while maintaining the same quality. In a separate test, engineering teams completed the same number of pull-requests with 25 percent greater token efficiency4.

Snowflake Introduces New Open Models to Governed Enterprise Data

DeepSeek-V4-Flash 0731 and GLM-5.3 are the latest models that will be added to Snowflake’s growing portfolio of models available across Cortex AI, Snowflake CoCo, and Snowflake CoWork. Together with models from providers including Anthropic, OpenAI, Google, SpaceXAI, Meta, and Mistral, this portfolio gives customers more freedom to choose the right combination of performance and cost for each workload.

Rather than treating each open model release as a standalone integration, Snowflake is building an environment designed to keep pace with the rapidly evolving open model ecosystem. Snowflake continuously evaluates and optimizes how these models are served and used across its AI products, helping customers take advantage of advances in the open model ecosystem without having to repeatedly rework their apps or infrastructure. As a result, customers can benefit from newer and more capable models, while applying consistent access controls and governance across both open and proprietary options.

Snowflake’s AI Research Team evaluated DeepSeek-V4-Flash on enterprise-focused tasks, with recent testing showing that DeepSeek-V4-Flash outperformed the leading proprietary models used for the evaluation, scoring 74.4 percent on data engineering tasks. GLM-5.2 also performed strongly at 62.8 percent, while using fewer tokens than any other model tested5. These results demonstrate that open models are increasingly capable of supporting mission-critical tasks at a lower cost.

Snowflake Helps Enterprises Improve Intelligence Efficiency at Scale

As enterprises use more models across a growing number of AI apps and agents, improving intelligence efficiency requires visibility and control over how AI resources are consumed. Cortex AI Gateway gives administrators visibility into token usage and costs, while allowing organizations to establish spending limits across AI apps and agents. Snowflake CoCo extends those controls through Snowflake’s existing role-based access and tagging framework, allowing administrators to set default models, attribute usage to teams or cost centers, establish per-user quotas, and receive notifications as consumption approaches defined limits.

Coupled with dynamic model routing and expanded model choice, these capabilities give enterprises greater control over their AI economics. Organizations can automatically match workloads to the right models, take advantage of new model options as they emerge, and manage how AI resources are consumed across the business. This helps enterprises improve intelligence efficiency while maintaining Snowflake’s trusted governance and retaining ownership of their proprietary enterprise data and context.

Learn More:

  • To learn more about how enterprises can maximize intelligence efficiency, read Snowflake CEO Sridhar Ramaswamy’s blog post.

  • Double click into how Snowflake’s latest AI innovations are driving better AI economics in this blog post.

  • Stay on top of the latest news and announcements from Snowflake on LinkedIn and X.

1 Snowflake product is now in private preview.

2 Snowflake product will be private preview soon.

3 Snowflake product will be private preview soon, but is subject to change based on model availability.

4 Results based on Snowflake internal testing; methodology and conditions available upon request. Individual results may vary based on workload and configuration.

5 Efficiency score based on internal testing using ADE-bench, a framework created by dbt for evaluating AI agents on real-world analytics and data engineering tasks.

About Snowflake

Snowflake is the platform for the AI era, making it easy for enterprises to innovate faster and get more value from data. More than 13,900 customers around the globe, including hundreds of the world’s largest companies, use Snowflake’s AI Data Cloud to build, use and share data, applications and AI, in a secure and governed fashion. With Snowflake, data and AI are transformative for everyone. Learn more at snowflake.com (NYSE: SNOW).

Forward‑Looking Statements

This press release contains express and implied forward-looking statements, including statements regarding (i) our future operating results, targets, or financial position; (ii) our business strategy, plans, opportunities, or priorities; (iii) the release, adoption, and use of our new or enhanced products, services, and technology offerings, including those that are under development or not generally available; (iv) market size and growth, trends, and competitive considerations; (v) our vision, strategy and expected benefits relating to artificial intelligence (AI), the enterprise AI revolution, Snowflake Cortex AI, Snowpark, Snowflake Marketplace, the AI Data Cloud, and AI Data Clouds for specific industries or product categories, including the expected benefits and network effects of the AI Data Cloud; and (vi) the integration, interoperability, and availability of our products, services, and technology offerings with and on third-party products and platforms, including public cloud platforms and AI models. These forward-looking statements are subject to a number of risks, uncertainties and assumptions, including those described under the heading “Risk Factors” and elsewhere in the Quarterly Reports on Form 10-Q and the Annual Reports on Form 10-K that Snowflake files with the Securities and Exchange Commission. In light of these risks, uncertainties, and assumptions, actual results could differ materially and adversely from those anticipated or implied in the forward-looking statements. As a result, you should not rely on any forward-looking statements as predictions of future events.

© 2026 Snowflake Inc. All rights reserved. Snowflake, the Snowflake logo, and all other Snowflake product, feature and service names mentioned herein are registered trademarks or trademarks of Snowflake Inc. in the United States and other countries. All other brand names or logos mentioned or used herein are for identification purposes only and may be the trademarks of their respective holder(s). Snowflake may not be associated with, or be sponsored or endorsed by, any such holder(s).

Media gallery

About The Author
Business Wire