-

MIPS S8200 Delivers Software-First RISC-V NPU To Enable Physical AI at the Autonomous Edge

  • MIPS S8200 NPU is sampling now to lead customers developing autonomous edge transportation, robotics, and embedded platforms
  • ForwardEdge ASIC selects MIPS S8200 NPU for advanced autonomous mission-critical platforms
  • MIPS enables universal on-device AI capabilities with support for modern models and class-leading efficiency

SAN JOSE, Calif.--(BUSINESS WIRE)--MIPS, a GlobalFoundries company, today unveiled details of the MIPS S8200 processor IP to enable next-generation AI workloads in embedded platforms. The most recent addition to the MIPS Atlas family of RISC-V processor IP, the MIPS S8200 RISC-V NPU delivers support for transformer and agentic language AI models at the edge, increased efficiency, and performance improvements. This innovation marks a significant step forward in enabling standards-based, scalable Physical AI solutions for autonomous edge platforms across multiple markets.

“MIPS is bringing Physical AI to the autonomous edge platforms that need modern AI capabilities in a scalable and highly efficient way that is also open and standards based. MIPS offers increased performance and efficiency for our customers' autonomous platforms with the MIPS S8200 RISC-V NPU,” said Sameer Wasson, MIPS CEO. “MIPS virtual platform, Atlas Explorer, enables our customers to move fast and begin model optimization ahead of silicon. MIPS is enabling the autonomous edge platforms that will move AI from the datacenter into physical commercial and industrial products.”

ForwardEdge ASIC, a wholly owned, commercially focused subsidiary of Lockheed Martin and leader in trusted microelectronics solutions for mission-critical applications in aerospace, defense and other high-reliability industries, has selected the MIPS S8200 for an upcoming high-performance application-specific ASIC to be used in autonomous platforms.

“Our mission to build high-performance solutions for autonomous platforms requires class-leading performance and area density when running the latest AI models,” said Frank Ferrante, CEO ForwardEdge ASIC. “MIPS S8200 delivers the superior performance and efficiency we need for our advanced mission-critical platforms.”

The MIPS S8200 delivers the true multimodal intelligence needed to execute Physical AI in autonomous edge platforms such as diverse neural networks for vision, radar, speech and more, and sophisticated contextual awareness. MIPS’ software-first approach enables customers to begin model optimization on virtual platforms, using MIPS Atlas Explorer. These Core Models enable developers to co-design hardware and software, unlocking efficiency and application optimizations. Combined with the open, modular standards of RISC-V, MIPS enables lower platform development costs.

In 2027, MIPS expects to sample the first silicon reference platforms featuring the MIPS S8200 neural processing unit to accelerate the adoption of Physical AI in autonomous edge devices. These evaluation platforms will showcase the performance and efficiency of MIPS AI platforms and drive the ecosystem development of toolchains to support the adoption of RISC-V, including Vector and Matrix extensions, in autonomous edge applications.

To learn more about MIPS Physical AI solutions for autonomous edge platforms, visit MIPS at CES 2026, in the LVCC West Hall, Booth 3253; or Contact Us.

About MIPS

MIPS, a GlobalFoundries company, is a leading provider of RISC-V processor IP, software and tools for physical AI platforms. Our innovations deliver standards-based autonomous edge platforms developed with our customers’ software-first requirements. MIPS technology enables the adoption of AI in real-time, event-driven mission-critical platforms for high-growth markets such as transportation, robotics, and embedded computing. As part of the GlobalFoundries family, we deliver RISC-V at foundry scale and enable Physical AI to be built on MIPS. For more information visit MIPS.com.

Contacts

Media Contact
Tyler Weiland
Shelton Group
+1-972-571-7834
tweiland@sheltongroup.com

MIPS Contact
James Prior
MIPS Marketing
+1-518-974-5882
press@mips.com

MIPS


Release Versions

Contacts

Media Contact
Tyler Weiland
Shelton Group
+1-972-571-7834
tweiland@sheltongroup.com

MIPS Contact
James Prior
MIPS Marketing
+1-518-974-5882
press@mips.com

More News From MIPS

MIPS I8500 Processor Orchestrates Data Movement for the AI Era

SAN JOSE, Calif.--(BUSINESS WIRE)--MIPS, a GlobalFoundries company, announced today the MIPS I8500 processor is now sampling to lead customers. Featured at GlobalFoundries’ Technology Summit in Munich, Germany today, the I8500 represents a class of intelligent data movement processor IP designed for real-time, event-driven computing platforms. Targeting hyperscale, storage, automotive, industrial, and communications infrastructure markets, the I8500 is built to meet the demands of the AI superc...

MIPS Appoints Moshe Hillel to Accelerate Growth in Israel

SAN JOSE, Calif.--(BUSINESS WIRE)--MIPS, a GlobalFoundries company, announced today that Moshe Hillel has joined as head of Business Development for Israel. In this role, Moshe will drive MIPS RISC-V adoption across the region, building on his extensive background in semiconductor solutions for analog and digital embedded application processing to support MIPS mission to advance the next wave of physical AI. “The MIPS mission to drive intelligence into action with RISC-V continues to increase m...

MIPS Appoints Alan Li as Head of Business Development to Accelerate China Growth

SAN JOSE, Calif.--(BUSINESS WIRE)--MIPS, a GlobalFoundries company, announced today that Alan Li has joined as head of Business Development for China. In this role, Alan will lead business expansion across the region, building on his extensive background in automotive, industrial, and communications infrastructure sectors to support MIPS mission to advance the next wave of physical AI. “The adoption of physical AI at the edge and endpoint drives intense new requirements for computing platforms,...
Back to Newsroom