SambaNova and Intel have launched an inference architecture to support agentic AI workloads.

The offering will combine GPUs, SambaNova’s recently launched SN50 Reconfigurable Data Unit (RDU), and Intel Xeon 6 CPUs.

According to SambaNova, this architecture will allow AI inferencing to split into two phases: a compute intensive prompt processing, or ‘prefill,’ stage, and the memory bandwidth-intensive output generation stage, known as ‘decode.’

The GPUs will support prefill, while SambaNova’s RDUs will be used for decode. Intel’s CPUs function as the system control plane, supporting agentic task coordination, workload distribution, tool and API execution, and system‑level behavior, while also serving as the action CPU that compiles and executes code and validates results.

The CPUs will also provide the memory bandwidth, PCIe lane density, and on‑die accelerators. According to the chip and server firm, Intel’s Xeon 6 delivers more than 50 percent faster LLVM compilation times compared with Arm‑based server CPUs, and up to 70 percent faster vector database performance compared with available x86‑based offerings.

The design will be made available in H2 2026 to enterprises, cloud providers, and sovereign AI programs wanting to run coding agents and other agentic workloads at scale, SambaNova said.

“Agentic AI is moving into production - and the winning pattern we’re seeing is GPUs to start the job, Intel Xeon 6 to run it, and SambaNova RDUs to finish it fast,” said Rodrigo Liang, CEO and co‑founder of SambaNova Systems. “Together with Intel, we’re giving customers a blueprint they can deploy in existing air‑cooled data centers, with broad x86 coverage for the coding agents and tools they already use today.”

“The data center software ecosystem is built on x86, and it runs on Xeon—providing a mature, proven foundation that developers, enterprises, and cloud providers rely on at scale,” said Kevork Kechichian, EVP and GM of the Data Center Group (DCG) at Intel. “Workloads of the future will require a heterogeneous mix of computing, and this collaboration with SambaNova delivers a cost‑efficient, high‑performance inference architecture designed to meet customer needs at scale — powered by Xeon 6.”

The announcement comes a month after SambaNova unveiled its SN50 chip for agentic AI workloads, hardware that the company claims is 5x faster than “competitive chips” and offers a 3x lower total cost of ownership

The February launch also saw SambaNova announce a “multi-year strategic collaboration” with Intel to deliver “high‑performance, cost‑efficient AI inference solutions for AI‑native companies, model providers, enterprises, and government organizations around the world.”

In addition to the newly unveiled architecture, the two companies will also partner to roll out an Intel‑powered AI cloud built on Intel Xeon‑based infrastructure.