Cadence's Neo NPU Comes out on TOPS
October 19, 2023
San Jose. California. Cadence Design Systems introduced its innovative AI IP and software tools for on-device and edge AI processing. The scalable Cadence Neo Neural Processing Units (NPUs) support AI performance utilizing low energy while delivering up to 80 TOPS in a singular core. The NPUs offer both classic and new generative AI models for offloading AI/ML performance from any host processor (application processors, general-purpose microcontrollers, and DSPs) leveraging AMBA AXI interconnect.
Included is NeuroWeave Software Development Kit (SDK) enabling a no-code platform for all Cadence AI and Tensilica IP products. Flexibility options include Int4, Int8, Int16, and FP16 data types forming the foundation of CNN, RNN and transformer-based networks. Cadence designed the NPUs to perform at a high level without losing efficacy. The Neo NPUs have 20X the performance of previous generation with 2-5X the inferences per second per area (IPS/mm2 ) and 5-10X the inferences per second per Watt (IPS/W). The devices offer a broad configuration range of 256 to 32K MACs per cycle and simple integration of many network topologies and operators.
Ideal applications include intelligent sensors, IoT, audio/vision, hearables/wearables, mobile vision/voice AI, AR/VR, and ADAS.
David Glasco, Vice President of Research and Development for Tensilica IP at Cadence concludes, “In today’s rapidly evolving landscape, it’s critical that our customers are able to design and deliver AI solutions based on their unique requirements and KPIs without concern about whether future neural networks are supported. Toward this end, we’ve made significant investments in our new AI hardware platform and software toolchain to enable AI at every performance, power, and cost point and to drive the rapid deployment of AI-enabled systems.”
For more information, visit cadence.com.