Power consumption of Etos U55 NPU

Tobias Well, power consumption depends on the config/frequency/workloads. So, to give specific numbers, we need to know more details of what config/frequency. you are looking. It would be best to reach out to us at ml-support@arm.com for your specific request. 

Now, recently our partner Alif has released the U55 Silicon and some numbers which are public domain and might interest you:

for image classification (using MobileNetV2 model). According to Alif, the key results are:

• High-Efficiency (HE) subsystem for keyword spotting
  • Per inference, Ethos-U55 + Cortex-M55 is 28.9x faster and 33.0x more energy-efficient than Cortex-M55 alone
• High-Performance (HP) subsystem for image classification
  • Per inference, Ethos-U55 + Cortex-M55 is 75.2x faster and 76.0x more energy-efficient than Cortex-M55 alone
• As claimed by Arm, when both Ethos-U55 and Cortex-M55 are used together there is an 800x faster time-to-inference than compared to a MCU using a previous generation Cortex-M CPU with no microNPU.
  
  Source: https://www.embeddedcomputing.com/technology/ai-machine-learning/ai-logic-devices-worload-acceleration/alif-semiconductor-achieves-increased-power-performance-for-ai-ml-on-arm-ethos-u55-micronpu-and-cortex-m55-cpu-core

Tobias Well, power consumption depends on the config/frequency/workloads. So, to give specific numbers, we need to know more details of what config/frequency. you are looking. It would be best to reach out to us at ml-support@arm.com for your specific request. 

Now, recently our partner Alif has released the U55 Silicon and some numbers which are public domain and might interest you:

for image classification (using MobileNetV2 model). According to Alif, the key results are:

• High-Efficiency (HE) subsystem for keyword spotting
  • Per inference, Ethos-U55 + Cortex-M55 is 28.9x faster and 33.0x more energy-efficient than Cortex-M55 alone
• High-Performance (HP) subsystem for image classification
  • Per inference, Ethos-U55 + Cortex-M55 is 75.2x faster and 76.0x more energy-efficient than Cortex-M55 alone
• As claimed by Arm, when both Ethos-U55 and Cortex-M55 are used together there is an 800x faster time-to-inference than compared to a MCU using a previous generation Cortex-M CPU with no microNPU.
  
  Source: https://www.embeddedcomputing.com/technology/ai-machine-learning/ai-logic-devices-worload-acceleration/alif-semiconductor-achieves-increased-power-performance-for-ai-ml-on-arm-ethos-u55-micronpu-and-cortex-m55-cpu-core