In regions with unstable networks, small AI models are gradually becoming popular.

robot
Abstract generation in progress
ME News reports that on July 7 (UTC+8), 2019, Adebayo Alonge's RxScanner, whose servers were located in the U.S., took over 5 minutes per scan. Engineers reduced the AI model to run locally on an Android phone within 2 hours, after which RxScanner could verify drugs in areas without broadband or electricity. Small AI models typically have at most a few billion parameters, can run on phones or Raspberry Pi, and consume only a few watts of power. Similar cases include drones for detecting cashew diseases in India, ant invasion identification in Uruguay, malaria mosquito detection, and Arduino-based ECG devices in Brazil. A World Bank report shows that only 0.7% of internet users in the poorest countries have used ChatGPT, compared to a quarter in developed countries; the bank president believes small AI is key to providing life-saving services in regions lacking computing power and electricity. 🔗 Read the original article: via AI HOT ·
View Original
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • Comment
  • Repost
  • Share
Comment
Add a comment
Add a comment
No comments
  • Pinned