A few months after the MLPerft Tiny V1.4 benchmark by MLCommons and ASYGN COlibryNPU is still the world’s most energy-efficient NPU. ⚡
With a consumption of 22 µJ per inference for image recognition, Colibry is the NPU you need to integrate AI into your systems without changing the original battery. Actually, at one inference per second, a system integrating ColibryNPU can operate for over 3 years on a single CR2032 coin cell battery (220 mAh). 🔋
For more detail, the complete benchmark is available here:
https://lnkd.in/g8gv9cPt
⚙️ Colibry NPU specs:
<1mW Ai processing
32-bit RISC-V CPU
128 kB CPU SRAM Memory
9.6 GOPs compute power
300 MHz max frequency
640 kB NNPA SRAM Memory
ColibryNPU also works with different sensor data! Want to try out?
Contact us to receive your evaluation board, and let’s exchange about your use case! 🚀
https://lnkd.in/eUJ5B8bQ
