Baseten, the platform for mission-critical inference, announced today that it has signed a Strategic Collaboration Agreement (SCA) with Amazon Web Services, Inc. (AWS), expanding availability of Baseten’s inference services to customers deploying AI applications on AWS.

Baseten addresses a critical challenge for AI-driven enterprises: running large, custom AI models efficiently in production. The Baseten Inference Stack enables organizations to serve more AI requests per GPU with lower latency and higher throughput. This means engineering teams can focus on innovation rather than managing servers and optimization details. Through the SCA, Baseten and AWS will deepen their technical integration and joint go-to-market efforts. This partnership gives enterprises a secure, compliant way to use Baseten's high-performance inference technology on their own AWS infrastructure, keeping full control of their data and systems. Customers get easy setup, automatic scaling, and faster AI model performance, all while using their existing AWS contracts and resources.

"AWS provides the infrastructure foundation that enables us to deliver exceptional inference performance for our customers," said Tuhin Srivastava, CEO and Co-Founder at Baseten. "Through this collaboration, customers can combine Baseten's inference stack with the AWS secure global infrastructure to run their most demanding AI workloads while maintaining full control of their data and leveraging their existing AWS investments."

"Our partner network plays a critical role in helping customers unlock the value of AI,” said Chris Sullivan, Vice President, Americas Channels & Alliances at AWS. "We're committed to helping growth-stage companies scale their breakthrough technologies while simultaneously delivering transformative capabilities to their customers. Collaborations like this enable customers to combine AWS's secure, scalable infrastructure with specialized partner innovations—giving them the performance, flexibility, and confidence they need to deploy AI applications at scale.”

Baseten's hybrid deployment model empowers them to run workloads on their dedicated AWS VPC environments, seamlessly extending to Baseten Cloud during demand. This flexible architecture leverages AWS’s robust and scalable foundation, while adding an extra layer of agility, providing consistent performance, enhanced resilience, and secure, uninterrupted scaling exactly when needed.

Scaled Cognition, creators of the Agentic Pretrained Transformer (APT-1) and leaders in high-performance agentic AI workflows, exemplifies how Baseten's solution delivers value for AI-native companies on AWS. The company needed to serve APT-1 with ultra-low latency for mission-critical workloads while maintaining enterprise security and maximizing their existing AWS investments. By deploying Baseten's inference stack on their own AWS GPUs within their VPC, Scaled Cognition achieved time-to-first-token under 120ms and reduced overall latency by 40%.

“We really appreciate the collaboration with Baseten. The hybrid cloud solution and access to cutting-edge GPUs while working with our existing AWS commitments were key to our success on launch day and beyond. Beyond that, Baseten’s developer experience has been a favorite across our teams” — said Jordan DeLoach, VP of Engineering, Scaled Cognition

This collaboration underscores the shared commitment of Baseten and AWS to provide flexibility, performance, and business value for customers developing and deploying AI solutions across industries.

For more information, please visit https://www.baseten.co/

About Baseten

Baseten is the leader in inference for high-scale AI products, providing the industry’s most advanced inference stack. Purpose-built for performance, reliability, and cost-efficiency, Baseten enables teams to build and scale the next great AI products. Through applied research, production-grade infrastructure, and a seamless developer experience, customers can infinitely scale open-source, custom, and fine-tuned models in production. Backed by BOND, CapitalG, IVP, Spark, Greylock, Conviction, and others, Baseten is trusted by the fastest-growing AI companies to power their most ambitious products. Learn more at www.baseten.co

View source version on businesswire.com: https://www.businesswire.com/news/home/20251203239212/en/