Ship our white-label app with your brand in weeks, or add the SDK to the app you already have. Either way, updates and reporting are handled for you.
Your name, logo and colours on Android, iOS, Windows and Mac. Reporting, updates and offline answers are built in, so there’s no app to write.
Kotlin, Swift, TypeScript, C++ and Python, with the same API everywhere. Load a model, stream an answer, switch on reporting.
Load the model your team trained, stream the answer word by word, and turn on reporting. Updates arrive without another line of code.
// Load the model your team trained. New versions arrive over the air.val ai = RageEdge.load("service-assistant") // Stream the answer word by word, fully offline.ai.ask("Torque for the front caliper bolts?").collect { token -> answerView.append(token)} // Add a Report button that routes to the right team.ai.enableReporting(team = "service")// Load the model your team trained. New versions arrive over the air.let ai = try await RageEdge.load("service-assistant") // Stream the answer word by word, fully offline.for try await token in ai.ask("Torque for the front caliper bolts?") { answerView.append(token)} // Add a Report button that routes to the right team.ai.enableReporting(team: "service")import { RageEdge } from "@rageedge/sdk"; // Load the model your team trained. New versions arrive over the air.const ai = await RageEdge.load("service-assistant"); // Stream the answer word by word, fully offline.for await (const token of ai.ask("Torque for the front caliper bolts?")) { answer.append(token);} ai.enableReporting({ team: "service" });Code shown is illustrative. Package names and signatures may change before general availability.
RageEdge.load(name)Opens the newest signed version on the device. Works offline from the first launch.
ai.ask(question)Streams the answer token by token, with the source page it came from.
ai.enableReporting(team)Adds the Report button and routes reports to the team you name.
ai.onUpdate(callback)Tells your app when a new version is ready. Swapping happens when the app is idle.
One model family, built for each device’s chip. Pick the white-label app, the SDK, or both.
| Platform | White-label app | SDK | Acceleration |
|---|---|---|---|
| Android | Yes | Kotlin | NPU, GPU or CPU |
| iOS and iPadOS | Yes | Swift | Neural Engine, GPU |
| Windows | Yes | C++, TypeScript | GPU or CPU |
| macOS | Yes | Swift, TypeScript | Apple silicon |
| Linux and IoT | Kiosk build | C++, Python | GPU, NPU or CPU |
| In-car | On request | C++, Kotlin | Vehicle SoC |
Device support depends on memory and chip. We confirm your exact devices during the pilot.
The SDK and the white-label app follow the same safe path for every new version.
The app asks for a new version when it has a connection.
Only what changed, in the background, on Wi-Fi if you prefer.
The signature is checked before anything is loaded.
The new model takes over when the app is idle. No restart.
If you roll back in the console, devices return to the previous version.
Set the policy per device group in the console, or override it in code. The SDK follows it on every platform.
Every time the app opens, once a day, once a week, or the moment the device comes back online.
Keep mobile data for work. Only the part of the model that changed is downloaded.
After a set number of days without a check, the app asks the user to connect once so no device drifts too far behind.
Answer on the device, use your larger cloud model whenever there’s a connection, or stay online for always-connected desks and kiosks.
Field phones, service-bay tablets and line gateways can each follow their own rules.
Devices in this group check as soon as they’re online, download on Wi-Fi only, and connect at least every 14 days. Answers always come from the device.
ai.updatePolicy( check = Check.WHEN_ONLINE, // or ON_APP_OPEN, DAILY, WEEKLY network = Network.WIFI_ONLY, maxDaysOffline = 14, // then ask the user to connect once mode = Mode.AUTO // OFFLINE, AUTO or ONLINE)Code shown is illustrative.
Pick one team and one use case. We’ll have it running on your own devices in weeks.
Ask ChatGPT, Claude or Perplexity and see what they think. One click, instant insight.