Dart Mobile AIFlutterOn-DeviceMultimodal
llamafu
Run AI models directly on mobile devices. No cloud. No latency. Complete privacy. A Flutter FFI plugin for on-device LLM inference.
Key Features
📱
100% On-Device
No API keys, no network calls. Works completely offline with full privacy.
👁️
Vision & Multimodal
Support for LLaVA, Qwen2-VL and other vision models. Process images and audio locally.
🔧
Tool Calling
Function calling and structured JSON output with schema validation.
🔄
LoRA Hot-Swapping
Load and swap LoRA adapters at runtime for domain-specific fine-tuning.
⚡
Streaming Support
Real-time token streaming for responsive chat interfaces.
🌍
Cross-Platform
Android (API 21+) and iOS (12.0+) with optimized native code via FFI.
Quick Start
flutter pub add llamafu
import 'package:llamafu/llamafu.dart';
void main() async {
final llamafu = await Llamafu.init(
modelPath: '/path/to/model.gguf',
threads: 4,
contextSize: 2048,
);
final result = await llamafu.complete(
prompt: 'Explain quantum computing in simple terms:',
maxTokens: 256,
temperature: 0.7,
);
print(result);
llamafu.close();
}
Vision / Multimodal
final llamafu = await Llamafu.init(
modelPath: '/path/to/llava-model.gguf',
mmprojPath: '/path/to/mmproj.gguf',
);
final result = await llamafu.multimodalComplete(
prompt: 'Describe this image:',
mediaInputs: [
MediaInput(type: MediaType.image, data: '/path/to/image.jpg'),
],
maxTokens: 200,
);
Tool Calling
final result = await llamafu.chatComplete(
messages: [
ChatMessage.user('What is the weather in Tokyo?'),
],
tools: [
Tool(
name: 'get_weather',
description: 'Get weather for a location',
parameters: {
'type': 'object',
'properties': {
'location': {'type': 'string'},
},
},
),
],
);
Why On-Device?
| Factor | Cloud API | llamafu (On-Device) |
|---|---|---|
| Privacy | Data sent to servers | Data never leaves device |
| Latency | 200-2000ms round trip | Sub-100ms |
| Cost | Per-token billing | One-time compute |
| Offline | Requires internet | Works anywhere |
| Control | Vendor-dependent | You own the stack |
Platform Requirements
| Platform | Minimum Version |
|---|---|
| Flutter | 3.10.0+ |
| Dart SDK | 3.1.0+ |
| Android | API 21+ (Android 5.0), NDK 21+ |
| iOS | 12.0+, Xcode 14+ |
Models must be in GGUF format. Quantized versions (Q4_K_M, Q8_0) recommended for mobile.