---
title: "Edge or cloud inference? A decision checklist"
canonical: "https://www.kies.com.cn/en/kb/edge-vs-cloud-inference"
lang: "en"
type: "knowledge-base"
publisher: "Shanghai Paixu Network Technology Co., Ltd."
pricing: "mixed"
kb_type: "guide"
alternate: "https://www.kies.com.cn/zh/kb/edge-vs-cloud-inference.md"
---

# Edge or cloud inference? A decision checklist

> Source (canonical): https://www.kies.com.cn/en/kb/edge-vs-cloud-inference
> Published by: Shanghai Paixu Network Technology Co., Ltd. — B2B supplier of AI devices and robotics
> Type: Guide
> Language: English · Also available in Chinese: https://www.kies.com.cn/zh/kb/edge-vs-cloud-inference.md

## Answer

If any of bandwidth, latency or privacy constrains you, run inference at the edge: millisecond response, video stays on-site, works offline. Cloud suits fast-iterating models and bursty compute; most factory solutions combine both.

## Details

Ask three questions: can you afford the bandwidth to stream video to the cloud? Does the control loop need millisecond response? Must data stay on-site? One yes puts inference on an edge device - which is exactly why Jetson- and RK3588-class hardware exists.

Conversely, weekly-retrained models, bursty compute or centrally managed fleets favour cloud or an on-prem cluster with L4/L40S nodes. Most lines combine both: real-time verdicts at the edge, iteration and data return in the cloud.

## Enquiries and technical support

Send the application — deployment environment, compute and interface needs, quantity and timing. An engineer replies with a selection proposal and a written quotation within one business day.

- Enquiry form: https://www.kies.com.cn/en/contact
- Email: ai@kies.com.cn
- Phone: 021-80394405