Expedera Collaborates with Cadence to Expand Edge AI Capabilities

Expedera’s Origin™ Evolution NPU pairs with the Cadence Tensilica NeuroEdge AICP to create a validated AI subsystem IP for higher performance, lower power, and reduced complexity in edge inference

Santa Clara, California — November 9, 2026 — Expedera, Inc., a leading provider of customizable neural processing unit (NPU) semiconductor IP, today announced a collaboration with Cadence to demonstrate an AI subsystem pairing Expedera’s Origin™ Evolution NPU with the Cadence Tensilica NeuroEdge 130 AI Co-Processor (AICP) in a pre-silicon simulation. The simulation demonstrates AI workload distribution and execution across both IP blocks. The combined AICP-NPU solution is designed to improve performance and power efficiency and reduce latency, cost, and complexity in edge AI inference systems.

Highlights

• Expedera has paired its customizable NPU IP with the Cadence Tensilica NeuroEdge 130 AICP in a pre-silicon simulation demonstrating an AI subsystem, using an approach that can be extended to other Tensilica DSPs.
• The solution combines Expedera’s NPU architecture with the Cadence Tensilica NeuroEdge 130 AICP’s scalable AI compute capabilities to support inference designs across edge nodes, smartphones, automotive systems, and data center applications.
• Expedera’s Origin Evolution NPU IP was recently named Best Edge AI Processor IP in the 2026 Edge AI and Vision Product of the Year Awards.

“Pairing Expedera’s packet-based NPU architecture with Cadence’s Tensilica NeuroEdge 130 AICP gives developers a validated way to add scalable, efficient AI acceleration to the workflows they already use,” said Sharad Chole, chief scientist at Expedera. “With models running on the combined AICP-NPU subsystem and agentic AI workflows helping us deliver customized IP more quickly, customers have a faster, lower-risk path to production edge AI tailored to their performance and power requirements.”

“As AI inference moves from cloud to edge, SoC providers need a complete and energy-efficient solution to run the latest AI workloads end-to-end,” said Amol Borkar, group director of product management and marketing, Tensilica Products Group at Cadence. “The Tensilica NeuroEdge 130 AICP perfectly complements Expedera’s customizable NPU IP, assisting with pre/post-processing and operator fallback while executing cutting-edge AI networks. This AICP-NPU combination gives customers a validated path to add dedicated AI acceleration that is both configurable and future-ready.”

Explore Expedera IP:

• NPU IP Core for Edge
• Neural engine IP - AI Inference for the Highest Performing Systems
• NPU IP Core for Mobile

Learn More

• Visit expedera.com and cadence.com for additional details.
• Meet the teams at ICCAD-Expo 2026, Beijing, November 19–20 ; visit Expedera at booth F091.

About Expedera

Expedera provides customizable neural engine semiconductor IP that dramatically improves performance and power efficiency while reducing latency, cost and complexity in edge AI inference applications. Its patented, packet-based architecture eliminates the inefficiencies of conventional layer-based NPU designs, delivering superior utilization without sacrificing model accuracy. Deployed in more than 10 million consumer devices, Expedera’s Origin™ Evolution Neural Processing Unit (NPU) solutions are proven across a wide range of applications, from edge nodes and smartphones to automotive and data centers. Headquartered in Santa Clara, California, Expedera supports customers worldwide through engineering and customer support offices across the UK, China, Japan, Taiwan, and Singapore. Learn more at expedera.com.