On-device AI Models
 | This article appears to have been generated by a large language model (such as ChatGPT) without having been rigorously scrutinized for verifiability, neutrality, original research, and copyright compliance. It may include misleading or inaccurate claims and fake references that sound plausible. (January 2026) (Learn how and when to remove this template message) |
On-device AI models are artificial intelligence systems that are designed, trained, and deployed to perform data processing and inference directly on edge or terminal devices, such as smartphones, drones, autonomous vehicles, and IoT sensors, without requiring data transmission to remote cloud servers. [1] This approach emphasizes real-time performance, operation under hardware resource constraints, and enhanced data privacy. [1]
Definition and characteristics
On-device AI models are distinguished by three primary characteristics:
- Real-time performance: They enable immediate response to user inputs or environmental changes, making them suitable for latency-sensitive applications. [1]
- Resource constraints: These models must operate within limited computational power, memory, and energy budgets, often necessitating algorithmic and architectural optimizations. [1]
- Data privacy: By keeping data local to the device, they reduce exposure to external networks and mitigate risks associated with data breaches during transmission. [1]
References
- ↑ 1.0 1.1 1.2 1.3 1.4 Wang, Xubin; Tang, Zhiqing; Guo, Jianxiong; Meng, Tianhui; Wang, Chenhao; Jia, Weijia (April 2025). "Empowering Edge Intelligence: A Comprehensive Survey on On-Device AI Models". [ACM Computing Surveys]. 57 (9): [1–39]. arXiv:2503.06027. doi:10.1145/3724420.
This article "On-device AI Models" is from Wikipedia. The list of its authors can be seen in its historical and/or the page Edithistory:On-device AI Models. Articles copied from Draft Namespace on Wikipedia could be seen on the Draft Namespace of Wikipedia and not main one.
