Edge AI: what it means.
Edge AI places AI applications close to the environments producing their input data. NVIDIA describes inference running in deployed devices after a model has been trained. Sensors, local computation and application software form the deployed system. (NVIDIA: edge AI)
Local processing and cloud use are not mutually exclusive. A system can infer near the data while using remote resources for training or updates. Decide which parts run where rather than assuming that edge means permanently disconnected. (NVIDIA: edge AI)
Put the definition to work.
In a robot example, a model can analyze sensor data locally. “Edge” describes where computation happens; it does not identify one chip type or mean the system never connects to the cloud. (NVIDIA: edge AI)
Questions to ask next.
Use these questions as a reading checklist, not as a device-selection recommendation. Exact manufacturer documentation takes priority over a general definition.
- Where inference and training run
- Local compute and memory needs
- Connectivity and update requirements
For the broader explanation, explore the electronics inside robots. This page defines the term; the companion guide connects it to the rest of the system.
Common Edge AI questions.
Does edge AI require an NPU?
No particular chip type is implied by the location label. Evaluate the workload and the available processor, accelerator and software support. (NVIDIA: edge AI)