# What is on-device AI?

On-device AI runs models inside your phone, tablet, or PC — enabling private, low-latency features without constant cloud calls.

On-device AI means the model executes on the same hardware you are holding or typing on — not on a remote server.

Apple Intelligence, Google Gemini Nano, and Windows Copilot+ PC features are consumer examples: transcription, photo search, and writing help that work with minimal or no cloud dependency.

On-device AI is the phone-and-laptop slice of localized AI. Developers ship it through Apple Core ML, ONNX Runtime, Qualcomm AI Engine SDK, and local runtimes like MLC LLM.



## Sources
1. [Apple — Apple Intelligence](https://www.apple.com/apple-intelligence/)
2. [Google DeepMind — Gemini Nano](https://deepmind.google/technologies/gemini/nano/)
3. [Microsoft — Copilot+ PCs](https://www.microsoft.com/en-us/windows/copilot-plus-pcs)
4. [Apple — Core ML documentation](https://developer.apple.com/documentation/coreml)
5. [ONNX Runtime — execution providers](https://onnxruntime.ai/docs/execution-providers/)
6. [Qualcomm — AI Engine Direct SDK](https://www.qualcomm.com/developer/software/qualcomm-ai-engine-direct-sdk)
7. [MLC LLM](https://llm.mlc.ai/)

---
Page: https://localizedaiguide.com/q/what-is-on-device-ai