SambaNova와 Supermicro가 SambaNova RDU와 GPU를 결합한 AI 추론 시스템 구축을 위한 전략적 협력을 발표했다. 초기 계획은 SN50 RDU 10,000개 규모다.

SAN JOSE, Calif., Nov. 4, 2026 — SambaNova has announced that Supermicro is entering into a strategic collaboration, initially planning for 10,000 SN50 RDU chips, to build and sell AI disaggregated inference systems that include RDUs and GPUs.

"Agentic AI is the next wave of inference, and it has to run everywhere," said Rodrigo Liang, co-founder and CEO of SambaNova. "Supermicro builds AI infrastructure for data centers globally. Now customers can order SN50 RDUs alongside GPUs, from a world-leading partner they trust. That's how fast, efficient inference gets to scale."

Agents work nonstop, and how fast they act depends on how fast the model responds. Supermicro's systems pair NVIDIA GPUs, which read the prompt, with SambaNova SN50 chips built to write the response quickly.

"Supermicro is committed to delivering flexible, high-performance AI infrastructure that supports the evolving requirements of our customers. Our collaboration with SambaNova expands our broad DCBBS portfolio by enabling heterogeneous architectures that integrate GPU computing with specialized inference acceleration, helping customers deploy efficient, scalable AI solutions," said Charles Liang, president and CEO of Supermicro.

Built, Sold and Supported by Supermicro

Supermicro will integrate, test, deliver and support the systems for customers, including providing on-site service. SambaNova engineers will get models running and tuned on SN50 and help Supermicro with deployments. The companies will start with NeoClouds and sovereign clouds, then expand to enterprises that run AI in their own data centers.

About SambaNova RDU Chips

SambaNova's SN50 is the company's fifth-generation Reconfigurable Dataflow Unit (RDU) chip, purpose-built for high-performance inference. RDUs serve the largest models at high speed and high throughput without requiring extreme rack power density, which makes them much easier to deploy in existing data centers. RDUs also work alongside existing NVIDIA GPUs to drastically improve system performance and efficiency, with the GPUs running the prefill phase and the RDUs running the decode phase. Operators can therefore add inference capacity without waiting for next-generation data center build-outs, which are constrained by power and cooling, and can extend the useful life of the GPUs they already own. As a result, SambaNova and Supermicro can deliver substantially more high-performance inference capacity in 2027 than would otherwise be possible.

About SambaNova

SambaNova is a leader in next-generation AI infrastructure, providing a full stack platform that powers high-performance inference for enterprises, NeoClouds, AI labs and service providers, and sovereign AI initiatives worldwide. Founded in 2017 and headquartered in San Jose, Calif., SambaNova delivers chips, systems, and cloud services that enable customers to deploy state-of-the-art models with superior performance, lower total cost of ownership, and rapid time to value.

Source: SambaNova