SDKsJuly 14, 2026
Android to iOS in Six Weeks, with the RunAnywhere SDK
A solo developer ported his $2K MRR local-AI app to iOS in six weeks with the RunAnywhere SDK.
Open Source — SDKs · Infra · Console
Open sourceEverything above
the kernel is open.
One C++ core — inference, model management, routing — with SDKs for Swift, Kotlin, React Native, Flutter, TypeScript, and C++. LLM, speech, vision, and voice on device by default, cloud by policy, in a few lines of code.
The stack
One C++ core. Every target falls out of it.
The SDKs are thin bindings over the same open-source runtime, so every platform gets the same models, the same API, and the same speed — and a new target is a new binding, not a rewrite.
the runanywhere stack
your app · sdks
swift
ios · macos
kotlin
android · jvm
react-native
ios · android
flutter
ios · android
typescript
web · wasm
c++ · rcli
cli · embedded
one core
runanywhere-core
C++ · shared by every binding
llmsttttsvlmembeddingsmodel mgmtroutingtelemetry
execution
MetalRT
apple silicon gpu
QHexRT
qualcomm hexagon npu
CPU runtime
every device
Cloud
by policy · any provider
targets
iosandroidmacoswindowslinuxwebembedded
The hybrid layer
On-device by default. Cloud by policy.
Every request starts local. When a task needs more than the device can give — a bigger model, a longer context — the Control Plane's routing policies send it to cloud and bring it back, without your app code changing. Latency, privacy, and cost decide; you set the policy.
latencyprivacycost
sub-10ms · stay localFor teams
Ship with the SDK.
Manage with the Console.
Fleet dashboard, OTA model updates, policy-based routing, and inference analytics. Manage models on thousands of devices without app-store releases.
fleet dashboardOTA model updatespolicy-based routinginference analytics
Four lines of code. Any modality.
Services
LLM
Language
STT
Speech
TTS
Voice
VLM
Vision
1 import RunAnywhere
2 let response = try await RunAnywhere.chat("Explain quantum computing")
3 print(response)
4 // Streaming: RunAnywhere.generateStream(prompt)
2 let response = try await RunAnywhere.chat("Explain quantum computing")
3 print(response)
4 // Streaming: RunAnywhere.generateStream(prompt)
Explain quantum computing in simple terms
LLM
Quantum computing uses quantum mechanics to process information in fundamentally different ways than classical computers, enabling exponentially faster calculations for specific problems.
Cost-effective AI
3-minute setup
Privacy-focused
Built with the SDK
Real apps, shipped on the public SDK.
runanywhere-sdksRCLImetalrt-binariesqmv.metalattention_decode.metalrms_norm.metalswiftkotlinreact-nativefluttertypescriptrcliwhisperpiper
Tutorials
Chat, STT, TTS, and a full voice pipeline.
From the blog
SDK engineering notes.
SDKs
SDKsMarch 13, 2026
How RunAnywhere SDK Powers On-Device AI Coaching in PickleRite
A deep-dive into how PickleRite — a pickleball performance tracker — runs a specialized LLM entirely on-device using RunAnywhere SDK. Zero cloud costs, full offline support, complete privacy.
SDKsFebruary 19, 2026
I Tried Running an LLM on a $150 Android Phone. Here's What Actually Happened.
And the rabbit hole that taught me more about Android internals than 3 years of app development.
Try it now