Build. Train. Optimize. Tiny AI on Chip
Build datasets, train models, and optimize libraries to run on ultra low-power MCUs locally
*An illustration of AI on different platform
Ignite the blaze of Tiny AI applications
Billions of IoT devices around the world based on microcontrollers (MCU): They are low-cost, low-power, small, almost everywhere in our lives. However, deploying and training Tiny AI models on MCU locally is extremly hard : No DRAM, no OS and strict memory constraint. That's the reason why most of these MCU are designed to be Interent of Things.
We have reservations about everything being connected to the Internet. This notion raises concerns about energy efficiency, Internet accessibility, associated costs, and user privacy. We are passionate about disconnecting the 'Internet' from IoT, allowing us to harness AI capabilities in our everyday devices without the necessity of connectivity.
As ONCHIP.AI, a tiny AI edge-computing technology company, we are happy to share our technology with other manufacturers to help them improve their products, committed to providing better experience with AI capabilities. We have no intention of monopolizing technology and thus uphold an open-minded attitude. If you are interested, please contact us through hello@onchip.ai 如有興趣，請聯繫 hello@onchip.ai ご興味がございましたら、お問い合わせください hello@onchip.ai