[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-10802":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":16,"stars7d":17,"stars30d":18,"stars90d":16,"forks30d":16,"starsTrendScore":17,"compositeScore":19,"rankGlobal":10,"rankLanguage":10,"license":20,"archived":21,"fork":21,"defaultBranch":22,"hasWiki":21,"hasPages":23,"topics":24,"createdAt":10,"pushedAt":10,"updatedAt":45,"readmeContent":46,"aiSummary":47,"trendingCount":16,"starSnapshotCount":16,"syncStatus":48,"lastSyncTime":49,"discoverSource":50},10802,"runanywhere-sdks","RunanywhereAI\u002Frunanywhere-sdks","RunanywhereAI","Production ready toolkit to run AI locally","https:\u002F\u002Fwww.runanywhere.ai",null,"C++",10335,359,30,33,0,3,22,42.67,"Other",false,"main",true,[25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44],"android","apple-intelligence","cpp","diffusion-models","edge","flutter","inference","ios","kotlin","llamacpp","llm","multimodal","ollama","on-device-ai","react-native","swift","vlm","voice-ai","web","websdk","2026-06-12 02:02:26","\u003Cp align=\"center\">\n  \u003Cimg src=\"examples\u002Flogo.svg\" alt=\"RunAnywhere Logo\" width=\"140\"\u002F>\n\u003C\u002Fp>\n\n\u003Ch1 align=\"center\">RunAnywhere\u003C\u002Fh1>\n\n\u003Cp align=\"center\">\n  \u003Cstrong>On-device AI for every platform.\u003C\u002Fstrong>\u003Cbr\u002F>\n  Run LLMs, speech-to-text, and text-to-speech locally — private, offline, fast.\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fapps.apple.com\u002Fus\u002Fapp\u002Frunanywhere\u002Fid6756506307\">\n    \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FApp_Store-Download-0D96F6?style=for-the-badge&logo=apple&logoColor=white\" alt=\"Download on App Store\" \u002F>\n  \u003C\u002Fa>\n   \n  \u003Ca href=\"https:\u002F\u002Fplay.google.com\u002Fstore\u002Fapps\u002Fdetails?id=com.runanywhere.runanywhereai\">\n    \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FGoogle_Play-Download-34A853?style=for-the-badge&logo=google-play&logoColor=white\" alt=\"Get it on Google Play\" \u002F>\n  \u003C\u002Fa>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002FRunanywhereAI\u002Frunanywhere-sdks\u002Fstargazers\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FRunanywhereAI\u002Frunanywhere-sdks?style=flat-square\" alt=\"GitHub Stars\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"LICENSE\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-Apache%202.0-blue?style=flat-square\" alt=\"License\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fdiscord.gg\u002FN359FBbDVd\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDiscord-Join-5865F2?style=flat-square&logo=discord&logoColor=white\" alt=\"Discord\" \u002F>\u003C\u002Fa>\n\u003C\u002Fp>\n\n## See It In Action\n\n\u003Cdiv align=\"center\">\n\u003Ctable>\n  \u003Ctr>\n    \u003Ctd align=\"center\" width=\"50%\">\n      \u003Cimg src=\"docs\u002Fgifs\u002Ftext-generation.gif\" alt=\"Text Generation\" width=\"240\"\u002F>\u003Cbr\u002F>\u003Cbr\u002F>\n      \u003Cstrong>Text Generation\u003C\u002Fstrong>\u003Cbr\u002F>\n      \u003Csub>LLM inference — 100% on-device\u003C\u002Fsub>\n    \u003C\u002Ftd>\n    \u003Ctd width=\"40\">\u003C\u002Ftd>\n    \u003Ctd align=\"center\" width=\"50%\">\n      \u003Cimg src=\"docs\u002Fgifs\u002Fvoice-ai.gif\" alt=\"Voice AI\" width=\"240\"\u002F>\u003Cbr\u002F>\u003Cbr\u002F>\n      \u003Cstrong>Voice AI\u003C\u002Fstrong>\u003Cbr\u002F>\n      \u003Csub>STT → LLM → TTS pipeline — fully offline\u003C\u002Fsub>\n    \u003C\u002Ftd>\n  \u003C\u002Ftr>\n  \u003Ctr>\u003Ctd colspan=\"3\" height=\"30\">\u003C\u002Ftd>\u003C\u002Ftr>\n  \u003Ctr>\n    \u003Ctd align=\"center\" width=\"50%\">\n      \u003Cimg src=\"docs\u002Fgifs\u002Fimage-generation.gif\" alt=\"Image Generation\" width=\"240\"\u002F>\u003Cbr\u002F>\u003Cbr\u002F>\n      \u003Cstrong>Image Generation\u003C\u002Fstrong>\u003Cbr\u002F>\n      \u003Csub>On-device diffusion model\u003C\u002Fsub>\n    \u003C\u002Ftd>\n    \u003Ctd width=\"40\">\u003C\u002Ftd>\n    \u003Ctd align=\"center\" width=\"50%\">\n      \u003Cimg src=\"docs\u002Fgifs\u002Fvisual-language-model.gif\" alt=\"Visual Language Model\" width=\"240\"\u002F>\u003Cbr\u002F>\u003Cbr\u002F>\n      \u003Cstrong>Visual Language Model\u003C\u002Fstrong>\u003Cbr\u002F>\n      \u003Csub>Vision + language understanding on-device\u003C\u002Fsub>\n    \u003C\u002Ftd>\n  \u003C\u002Ftr>\n\u003C\u002Ftable>\n\u003C\u002Fdiv>\n\n---\n\n## What is RunAnywhere?\n\nRunAnywhere lets you add AI features to your app that run entirely on-device:\n\n- **LLM Chat** — Llama, Mistral, Qwen, SmolLM, and more\n- **Speech-to-Text** — Whisper-powered transcription\n- **Text-to-Speech** — Neural voice synthesis\n- **Voice Assistant** — Full STT → LLM → TTS pipeline\n\nNo cloud. No latency. No data leaves the device.\n\n---\n\n## SDKs\n\n| Platform | Status | Installation | Documentation |\n|----------|--------|--------------|---------------|\n| **Swift** (iOS\u002FmacOS) | Stable | [Swift Package Manager](#swift-ios--macos) | [docs.runanywhere.ai\u002Fswift](https:\u002F\u002Fdocs.runanywhere.ai\u002Fswift\u002Fintroduction) |\n| **Kotlin** (Android) | Stable | [Gradle](#kotlin-android) | [docs.runanywhere.ai\u002Fkotlin](https:\u002F\u002Fdocs.runanywhere.ai\u002Fkotlin\u002Fintroduction) |\n| **Web** (Browser) | Beta | [npm](#web-browser) | [SDK README](sdk\u002Frunanywhere-web\u002F) |\n| **React Native** | Beta | [npm](#react-native) | [docs.runanywhere.ai\u002Freact-native](https:\u002F\u002Fdocs.runanywhere.ai\u002Freact-native\u002Fintroduction) |\n| **Flutter** | Beta | [pub.dev](#flutter) | [docs.runanywhere.ai\u002Fflutter](https:\u002F\u002Fdocs.runanywhere.ai\u002Fflutter\u002Fintroduction) |\n\n---\n\n## Quick Start\n\n### Swift (iOS \u002F macOS)\n\n```swift\nimport RunAnywhere\nimport LlamaCPPRuntime\n\n\u002F\u002F 1. Initialize\nLlamaCPP.register()\ntry RunAnywhere.initialize()\n\n\u002F\u002F 2. Load a model\ntry await RunAnywhere.downloadModel(\"smollm2-360m\")\ntry await RunAnywhere.loadModel(\"smollm2-360m\")\n\n\u002F\u002F 3. Generate\nlet response = try await RunAnywhere.chat(\"What is the capital of France?\")\nprint(response) \u002F\u002F \"Paris is the capital of France.\"\n```\n\n**Install via Swift Package Manager:**\n\n```\nhttps:\u002F\u002Fgithub.com\u002FRunanywhereAI\u002Frunanywhere-sdks\n```\n\n[Full documentation →](https:\u002F\u002Fdocs.runanywhere.ai\u002Fswift\u002Fintroduction) · [Source code](sdk\u002Frunanywhere-swift\u002F)\n\n---\n\n### Kotlin (Android)\n\n```kotlin\nimport com.runanywhere.sdk.public.RunAnywhere\nimport com.runanywhere.sdk.public.extensions.*\n\n\u002F\u002F 1. Initialize\nLlamaCPP.register()\nRunAnywhere.initialize(environment = SDKEnvironment.DEVELOPMENT)\n\n\u002F\u002F 2. Load a model\nRunAnywhere.downloadModel(\"smollm2-360m\").collect { println(\"${it.progress * 100}%\") }\nRunAnywhere.loadLLMModel(\"smollm2-360m\")\n\n\u002F\u002F 3. Generate\nval response = RunAnywhere.chat(\"What is the capital of France?\")\nprintln(response) \u002F\u002F \"Paris is the capital of France.\"\n```\n\n**Install via Gradle:**\n\n```kotlin\ndependencies {\n    implementation(\"com.runanywhere.sdk:runanywhere-kotlin:0.16.1\")\n    implementation(\"com.runanywhere.sdk:runanywhere-core-llamacpp:0.16.1\")\n}\n```\n\n[Full documentation →](https:\u002F\u002Fdocs.runanywhere.ai\u002Fkotlin\u002Fintroduction) · [Source code](sdk\u002Frunanywhere-kotlin\u002F)\n\n---\n\n### React Native\n\n```typescript\nimport { RunAnywhere, SDKEnvironment } from '@runanywhere\u002Fcore';\nimport { LlamaCPP } from '@runanywhere\u002Fllamacpp';\n\n\u002F\u002F 1. Initialize\nawait RunAnywhere.initialize({ environment: SDKEnvironment.Development });\nLlamaCPP.register();\n\n\u002F\u002F 2. Load a model\nawait RunAnywhere.downloadModel('smollm2-360m');\nawait RunAnywhere.loadModel('smollm2-360m');\n\n\u002F\u002F 3. Generate\nconst response = await RunAnywhere.chat('What is the capital of France?');\nconsole.log(response); \u002F\u002F \"Paris is the capital of France.\"\n```\n\n**Install via npm:**\n\n```bash\nnpm install @runanywhere\u002Fcore @runanywhere\u002Fllamacpp\n```\n\n[Full documentation →](https:\u002F\u002Fdocs.runanywhere.ai\u002Freact-native\u002Fintroduction) · [Source code](sdk\u002Frunanywhere-react-native\u002F)\n\n---\n\n### Flutter\n\n```dart\nimport 'package:runanywhere\u002Frunanywhere.dart';\nimport 'package:runanywhere_llamacpp\u002Frunanywhere_llamacpp.dart';\n\n\u002F\u002F 1. Initialize\nawait RunAnywhere.initialize();\nawait LlamaCpp.register();\n\n\u002F\u002F 2. Load a model\nawait RunAnywhere.downloadModel('smollm2-360m');\nawait RunAnywhere.loadModel('smollm2-360m');\n\n\u002F\u002F 3. Generate\nfinal response = await RunAnywhere.chat('What is the capital of France?');\nprint(response); \u002F\u002F \"Paris is the capital of France.\"\n```\n\n**Install via pub.dev:**\n\n```yaml\ndependencies:\n  runanywhere: ^0.16.0\n  runanywhere_llamacpp: ^0.16.0  # LLM text generation\n  # runanywhere_onnx: ^0.16.0   # Add this if you need STT, TTS, or Voice features\n```\n\n[Full documentation →](https:\u002F\u002Fdocs.runanywhere.ai\u002Fflutter\u002Fintroduction) · [Source code](sdk\u002Frunanywhere-flutter\u002F)\n\n---\n\n### Web (Browser)\n\n```typescript\nimport { RunAnywhere, TextGeneration } from '@runanywhere\u002Fweb';\n\n\u002F\u002F 1. Initialize\nawait RunAnywhere.initialize({ environment: 'development' });\n\n\u002F\u002F 2. Load a model\nawait TextGeneration.loadModel('\u002Fmodels\u002Fqwen2.5-0.5b-instruct-q4_0.gguf', 'qwen2.5-0.5b');\n\n\u002F\u002F 3. Generate\nconst result = await TextGeneration.generate('What is the capital of France?');\nconsole.log(result.text); \u002F\u002F \"Paris is the capital of France.\"\n```\n\n**Install via npm:**\n\n```bash\nnpm install @runanywhere\u002Fweb\n```\n\n[Full documentation →](sdk\u002Frunanywhere-web\u002F) · [Source code](sdk\u002Frunanywhere-web\u002F)\n\n---\n\n## Sample Apps\n\nFull-featured demo applications demonstrating SDK capabilities:\n\n| Platform | Source Code | Download |\n|----------|-------------|----------|\n| iOS | [examples\u002Fios\u002FRunAnywhereAI](examples\u002Fios\u002FRunAnywhereAI\u002F) | [App Store](https:\u002F\u002Fapps.apple.com\u002Fus\u002Fapp\u002Frunanywhere\u002Fid6756506307) |\n| Android | [examples\u002Fandroid\u002FRunAnywhereAI](examples\u002Fandroid\u002FRunAnywhereAI\u002F) | [Google Play](https:\u002F\u002Fplay.google.com\u002Fstore\u002Fapps\u002Fdetails?id=com.runanywhere.runanywhereai) |\n| Web | [examples\u002Fweb\u002FRunAnywhereAI](examples\u002Fweb\u002FRunAnywhereAI\u002F) | Build from source |\n| React Native | [examples\u002Freact-native\u002FRunAnywhereAI](examples\u002Freact-native\u002FRunAnywhereAI\u002F) | Build from source |\n| Flutter | [examples\u002Fflutter\u002FRunAnywhereAI](examples\u002Fflutter\u002FRunAnywhereAI\u002F) | Build from source |\n\n---\n\n## Starter Examples\n\nMinimal starter projects to get up and running with RunAnywhere on each platform:\n\n| Platform | Repository |\n|----------|------------|\n| Kotlin (Android) | [RunanywhereAI\u002Fkotlin-starter-example](https:\u002F\u002Fgithub.com\u002FRunanywhereAI\u002Fkotlin-starter-example) |\n| Swift (iOS) | [RunanywhereAI\u002Fswift-starter-example](https:\u002F\u002Fgithub.com\u002FRunanywhereAI\u002Fswift-starter-example) |\n| Flutter | [RunanywhereAI\u002Fflutter-starter-example](https:\u002F\u002Fgithub.com\u002FRunanywhereAI\u002Fflutter-starter-example) |\n| React Native | [RunanywhereAI\u002Freact-native-starter-app](https:\u002F\u002Fgithub.com\u002FRunanywhereAI\u002Freact-native-starter-app) |\n\n---\n\n## Playground\n\nReal-world projects built with RunAnywhere that push the boundaries of on-device AI. Each one ships as a standalone app you can build and run.\n\n### [Android Use Agent](Playground\u002Fandroid-use-agent\u002F)\n\nA fully on-device autonomous Android agent that controls your phone. Give it a goal like \"Open YouTube and search for lofi music\" and it reads the screen via the Accessibility API, reasons about the next action with an on-device LLM (Qwen3-4B), and executes taps, swipes, and text input -- all without any cloud calls. Includes a Samsung foreground boost that delivers a 15x inference speedup, smart pre-launch via Android intents, and loop detection with automatic recovery. Benchmarked across four LLM models on a Galaxy S24. **[Full benchmarks](Playground\u002Fandroid-use-agent\u002FASSESSMENT.md)**\n\n### [On-Device Browser Agent](Playground\u002Fon-device-browser-agent\u002F)\n\nA Chrome extension that automates browser tasks entirely on-device using WebLLM and WebGPU. Uses a two-agent architecture -- a Planner that breaks down goals into steps and a Navigator that interacts with page elements -- with both DOM-based and vision-based page understanding. Includes site-specific workflows for Amazon, YouTube, and more. All AI inference runs locally on your GPU after the initial model download.\n\n### [Swift Starter App](Playground\u002Fswift-starter-app\u002F)\n\nA full-featured iOS app demonstrating the RunAnywhere SDK's core AI capabilities in a clean SwiftUI interface. Includes LLM chat with on-device language models, Whisper-powered speech-to-text, neural text-to-speech, and a complete voice pipeline that chains STT, LLM, and TTS together with voice activity detection. A good starting point for building privacy-first AI features on iOS.\n\n### [Linux Voice Assistant](Playground\u002Flinux-voice-assistant\u002F)\n\nA complete on-device voice AI pipeline for Linux (Raspberry Pi 5, x86_64, ARM64). Say \"Hey Jarvis\" to activate, speak naturally, and get responses -- all running locally with zero cloud dependency. Chains Wake Word detection (openWakeWord), Voice Activity Detection (Silero VAD), Speech-to-Text (Whisper Tiny EN), LLM reasoning (Qwen2.5 0.5B Q4), and Text-to-Speech (Piper neural TTS) in a single C++ binary.\n\n### [OpenClaw Hybrid Assistant](Playground\u002Fopenclaw-hybrid-assistant\u002F)\n\nA hybrid voice assistant that keeps latency-sensitive components on-device (wake word, VAD, STT, TTS) while routing reasoning to a cloud LLM via OpenClaw WebSocket. Supports barge-in (interrupt TTS by saying the wake word), waiting chimes for cloud response feedback, and noise-robust VAD with burst filtering. Built for scenarios where on-device LLMs are too slow but you still want private audio processing.\n\n---\n\n## Features\n\n| Feature | iOS | Android | Web | React Native | Flutter |\n|---------|-----|---------|-----|--------------|---------|\n| LLM Text Generation | ✅ | ✅ | ✅ | ✅ | ✅ |\n| Streaming | ✅ | ✅ | ✅ | ✅ | ✅ |\n| Speech-to-Text | ✅ | ✅ | ✅ | ✅ | ✅ |\n| Text-to-Speech | ✅ | ✅ | ✅ | ✅ | ✅ |\n| Voice Assistant Pipeline | ✅ | ✅ | ✅ | ✅ | ✅ |\n| Vision Language Models | ✅ | — | ✅ | — | — |\n| Model Download + Progress | ✅ | ✅ | ✅ | ✅ | ✅ |\n| Structured Output (JSON) | ✅ | ✅ | ✅ | 🔜 | 🔜 |\n| Tool Calling | ✅ | ✅ | ✅ | — | — |\n| Embeddings | — | — | ✅ | — | — |\n| Apple Foundation Models | ✅ | — | — | — | — |\n\n---\n\n## Supported Models\n\n### LLM (GGUF format via llama.cpp)\n\n| Model | Size | RAM Required | Use Case |\n|-------|------|--------------|----------|\n| SmolLM2 360M | ~400MB | 500MB | Fast, lightweight |\n| Qwen 2.5 0.5B | ~500MB | 600MB | Multilingual |\n| Llama 3.2 1B | ~1GB | 1.2GB | Balanced |\n| Mistral 7B Q4 | ~4GB | 5GB | High quality |\n\n### Speech-to-Text (Whisper via ONNX)\n\n| Model | Size | Languages |\n|-------|------|-----------|\n| Whisper Tiny | ~75MB | English |\n| Whisper Base | ~150MB | Multilingual |\n\n### Text-to-Speech (Piper via ONNX)\n\n| Voice | Size | Language |\n|-------|------|----------|\n| Piper US English | ~65MB | English (US) |\n| Piper British English | ~65MB | English (UK) |\n\n---\n\n## Repository Structure\n\n```\nrunanywhere-sdks\u002F\n├── sdk\u002F\n│   ├── runanywhere-swift\u002F          # iOS\u002FmacOS SDK\n│   ├── runanywhere-kotlin\u002F         # Android SDK\n│   ├── runanywhere-web\u002F            # Web SDK (WebAssembly)\n│   ├── runanywhere-react-native\u002F   # React Native SDK\n│   ├── runanywhere-flutter\u002F        # Flutter SDK\n│   └── runanywhere-commons\u002F        # Shared C++ core\n│\n├── examples\u002F\n│   ├── ios\u002FRunAnywhereAI\u002F          # iOS sample app\n│   ├── android\u002FRunAnywhereAI\u002F      # Android sample app\n│   ├── web\u002FRunAnywhereAI\u002F          # Web sample app\n│   ├── react-native\u002FRunAnywhereAI\u002F # React Native sample app\n│   └── flutter\u002FRunAnywhereAI\u002F      # Flutter sample app\n│\n├── Playground\u002F\n│   ├── swift-starter-app\u002F          # iOS AI playground app\n│   ├── on-device-browser-agent\u002F    # Chrome browser automation agent\n│   ├── android-use-agent\u002F          # On-device autonomous Android agent\n│   ├── linux-voice-assistant\u002F      # Linux on-device voice assistant\n│   └── openclaw-hybrid-assistant\u002F  # Hybrid voice assistant (on-device + cloud)\n│\n└── docs\u002F                           # Documentation\n```\n\n---\n\n## Requirements\n\n| Platform | Minimum | Recommended |\n|----------|---------|-------------|\n| iOS | 17.0+ | 17.0+ |\n| macOS | 14.0+ | 14.0+ |\n| Android | API 24 (7.0) | API 28+ |\n| Web | Chrome 96+ \u002F Edge 96+ | Chrome 120+ |\n| React Native | 0.74+ | 0.76+ |\n| Flutter | 3.10+ | 3.24+ |\n\n**Memory:** 2GB minimum, 4GB+ recommended for larger models\n\n---\n\n## Contributing\n\nWe welcome contributions. See our [Contributing Guide](CONTRIBUTING.md) for details.\n\n```bash\n# Clone the repo\ngit clone https:\u002F\u002Fgithub.com\u002FRunanywhereAI\u002Frunanywhere-sdks.git\n\n# Set up a specific SDK (example: Swift)\ncd runanywhere-sdks\u002Fsdk\u002Frunanywhere-swift\n.\u002Fscripts\u002Fbuild-swift.sh --setup\n\n# Run the sample app\ncd ..\u002F..\u002Fexamples\u002Fios\u002FRunAnywhereAI\nopen RunAnywhereAI.xcodeproj\n```\n\n---\n\n## Support\n\n- **Discord:** [Join our community](https:\u002F\u002Fdiscord.gg\u002FN359FBbDVd)\n- **GitHub Issues:** [Report bugs or request features](https:\u002F\u002Fgithub.com\u002FRunanywhereAI\u002Frunanywhere-sdks\u002Fissues)\n- **Email:** founders@runanywhere.ai\n- **Twitter:** [@RunanywhereAI](https:\u002F\u002Ftwitter.com\u002FRunanywhereAI)\n\n---\n\n## License\n\nApache 2.0 — see [LICENSE](LICENSE) for details.\n","RunAnywhere 是一个用于在本地设备上运行AI的工具包，支持多种平台。其核心功能包括完全在设备上运行的大规模语言模型（如Llama、Mistral等）、语音转文字、文字转语音以及完整的语音助手流程。采用C++开发，具备跨平台特性，适用于iOS、Android、Web等多种环境，并且提供Swift、Kotlin、React Native等多语言SDK。特别适合需要在无网络连接或对数据隐私有严格要求的情况下使用AI功能的应用场景，确保了用户数据的安全性和处理速度。",2,"2026-06-11 03:30:15","top_topic"]