# Baseten partners with Microsoft on open model inference

URL: https://decisioninsights.ai/baseten-partners-with-microsoft-on-open-model-inference/

Baseten and Microsoft have formed a partnership that brings Baseten's inference capabilities to Microsoft Foundry. The arrangement is now in public preview, offering enterprises on Azure a way to run open models while relying on Foundry's security and governance features.Production workloads have grown more complex, and enterprises increasingly select open models to gain control over cost, speed, customization, and compliance. The offering lets those organizations skip securing GPU capacity, managing deployments, or onboarding another vendor, and work from the Foundry platform where they already build and govern AI applications and agents.The offering includes a choice of frontier open models: DeepSeek V4.1 Flash 1017, GLM 5.3, GLM 5.3 Flash, and Kimi K3. Each model runs on Baseten's complete inference stack. For residency requirements, global and U.S. data zones are available, zero data retention applies, and Azure Policy controls allow administrators to restrict specific models.Usage is billed through Azure as first-party Foundry consumption and can draw on an existing Microsoft Azure Consumption Commitment. Teams pick pay-as-you-go for usage-based access or Provisioned Throughput Units for SLA-backed performance. After an enterprise opts in, developers locate Baseten models in the Foundry model catalog, choose a model and deployment type, and see deployments ready within seconds, with no prompts or responses stored.“The next wave of AI innovation will come from what developers build when powerful open models meet high-performance infrastructure,” said Steve Sweetman, VP of Product at Microsoft Foundry. “Bringing Baseten to Microsoft Foundry gives enterprises more choice and pairs Baseten’s optimized and reliable inference with the governance, controls, and observability within the Foundry platform, so developers can turn their most ambitious ideas into production applications. Together, we’re helping customers create value faster.”“It’s easy to forget how early broad adoption of AI is. Many enterprises are just now seeing how open models can deliver better efficiency, reliability, and ownership,” said Amir Haghighat, co-founder and CTO at Baseten. “Microsoft is one of the most trusted and respected names for enterprises, and we’re eager to help its customers move faster in building durable, owned intelligence.” General availability is expected in early 2027.
