react-native-executorch 0.5.8 → 0.5.10

package/README.md

@@ -9,6 +9,15 @@
 
 ![Software Mansion banner](https://github.com/user-attachments/assets/fa2c4735-e75c-4cc1-970d-88905d95e3a4)
 
+<p align="center">
+  <a href="https://github.com/software-mansion/react-native-executorch/blob/release/0.5/README.md">English</a>
+  <a href="https://github.com/software-mansion/react-native-executorch/blob/release/0.5/readmes/README_es.md">Español</a>
+  <a href="https://github.com/software-mansion/react-native-executorch/blob/release/0.5/readmes/README_fr.md">Français</a>
+  <a href="https://github.com/software-mansion/react-native-executorch/blob/release/0.5/readmes/README_cn.md">简体中文</a>
+  <a href="https://github.com/software-mansion/react-native-executorch/blob/release/0.5/readmes/README_pt.md">Português</a>
+  <a href="https://github.com/software-mansion/react-native-executorch/blob/release/0.5/readmes/README_in.md">हिंदी</a>
+</p>
+
 **React Native ExecuTorch** provides a declarative way to run AI models on-device using React Native, powered by **ExecuTorch** :rocket:. It offers out-of-the-box support for a wide range of LLMs, computer vision models, and more. Visit our [HuggingFace](https://huggingface.co/software-mansion) page to explore these models.
 
 **ExecuTorch**, developed by Meta, is a novel framework allowing AI model execution on devices like mobile phones or microcontrollers.
@@ -21,10 +30,11 @@ React Native ExecuTorch bridges the gap between React Native and native platform
 **Table of contents:**
 
 - [:yin_yang: Supported versions](#yin_yang-supported-versions)
-- [:robot: Ready-made models](#robot-ready-made-models)
 - [:books: Documentation](#books-documentation)
+- [:earth_africa: Real-World Example](#earth_africa-real-world-example)
 - [:llama: Quickstart - Running Llama](#llama-quickstart---running-llama)
-- [:calling: Examples](#calling-examples)
+- [:calling: Demo apps](#calling-demo-apps)
+- [:robot: Ready-made models](#robot-ready-made-models)
 - [:balance_scale: License](#balance_scale-license)
 - [:soon: What's next?](#soon-whats-next)
 
@@ -38,15 +48,17 @@ The minimal supported version are:
 > [!IMPORTANT]  
 > React Native Executorch supports only the [New React Native architecture](https://reactnative.dev/architecture/landing-page).
 
-## :robot: Ready-made models
-
-Our library has a number of ready-to-use AI models; a complete list is available in the documentation. If you're interested in running your own AI model, you need to first export it to the `.pte` format. Instructions on how to do this are available in the [Python API](https://pypi.org/project/executorch/).
-
 ## :books: Documentation
 
 Check out how our library can help you build your React Native AI features by visiting our docs:  
 https://docs.swmansion.com/react-native-executorch
 
+## :earth_africa: Real-World Example
+
+React Native ExecuTorch is powering [Private Mind](https://github.com/software-mansion-labs/private-mind), a privacy-first mobile AI app available on [App Store](https://apps.apple.com/gb/app/private-mind/id6746713439) and [Google Play](https://play.google.com/store/apps/details?id=com.swmansion.privatemind).
+
+<img width="2720" height="1085" alt="Private Mind promo" src="https://github.com/user-attachments/assets/b12296fe-19ac-48fc-9726-da9242700346" />
+
 ## :llama: **Quickstart - Running Llama**
 
 **Get started with AI-powered text generation in 3 easy steps!**
@@ -93,12 +105,12 @@ const handleGenerate = async () => {
 };
 ```
 
-## :calling: Examples
+## :calling: Demo apps
 
 We currently host a few example [apps](https://github.com/software-mansion/react-native-executorch/tree/main/apps) demonstrating use cases of our library:
 
 - `llm` - Chat application showcasing use of LLMs
-- `speech-to-text` - Whisper and Moonshine models ready for transcription tasks
+- `speech-to-text` - Whisper model ready for transcription tasks
 - `computer-vision` - Computer vision related tasks
 - `text-embeddings` - Computing text representations for semantic search
 
@@ -117,6 +129,10 @@ yarn expo run:< ios | android >
 > [!WARNING]  
 > Running LLMs requires a significant amount of RAM. If you are encountering unexpected app crashes, try to increase the amount of RAM allocated to the emulator.
 
+## :robot: Ready-made models
+
+Our library has a number of ready-to-use AI models; a complete list is available in the documentation. If you're interested in running your own AI model, you need to first export it to the `.pte` format. Instructions on how to do this are available in the [Python API](https://docs.pytorch.org/executorch/stable/using-executorch-export.html) and [optimum-executorch README](https://github.com/huggingface/optimum-executorch?tab=readme-ov-file#option-2-export-and-load-separately).
+
 ## :balance_scale: License
 
 This library is licensed under [The MIT License](./LICENSE).

package/lib/module/hooks/general/useExecutorchModule.js

@@ -7,9 +7,7 @@ export const useExecutorchModule = ({
   preventLoad = false
 }) => useModule({
   module: ExecutorchModule,
-  model: {
-    modelSource
-  },
+  model: modelSource,
   preventLoad
 });
 //# sourceMappingURL=useExecutorchModule.js.map

package/lib/module/hooks/general/useExecutorchModule.js.map

@@ -1 +1 @@
-{"version":3,"names":["ExecutorchModule","useModule","useExecutorchModule","modelSource","preventLoad","module","model"],"sourceRoot":"../../../../src","sources":["hooks/general/useExecutorchModule.ts"],"mappings":";;AAAA,SAASA,gBAAgB,QAAQ,wCAAwC;AAEzE,SAASC,SAAS,QAAQ,cAAc;AAOxC,OAAO,MAAMC,mBAAmB,GAAGA,CAAC;EAClCC,WAAW;EACXC,WAAW,GAAG;AACT,CAAC,KACNH,SAAS,CAAC;EACRI,MAAM,EAAEL,gBAAgB;EACxBM,KAAK,EAAE;IAAEH;EAAY,CAAC;EACtBC;AACF,CAAC,CAAC","ignoreList":[]}
+{"version":3,"names":["ExecutorchModule","useModule","useExecutorchModule","modelSource","preventLoad","module","model"],"sourceRoot":"../../../../src","sources":["hooks/general/useExecutorchModule.ts"],"mappings":";;AAAA,SAASA,gBAAgB,QAAQ,wCAAwC;AAEzE,SAASC,SAAS,QAAQ,cAAc;AAOxC,OAAO,MAAMC,mBAAmB,GAAGA,CAAC;EAClCC,WAAW;EACXC,WAAW,GAAG;AACT,CAAC,KACNH,SAAS,CAAC;EACRI,MAAM,EAAEL,gBAAgB;EACxBM,KAAK,EAAEH,WAAW;EAClBC;AACF,CAAC,CAAC","ignoreList":[]}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "react-native-executorch",
-  "version": "0.5.8",
+  "version": "0.5.10",
   "description": "An easy way to run AI models in React Native with ExecuTorch",
   "source": "./src/index.ts",
   "main": "./lib/module/index.js",

package/src/hooks/general/useExecutorchModule.ts

@@ -13,6 +13,6 @@ export const useExecutorchModule = ({
 }: Props) =>
   useModule({
     module: ExecutorchModule,
-    model: { modelSource },
+    model: modelSource,
     preventLoad,
   });

package/lib/Error.js

@@ -1,53 +0,0 @@
-export var ETError;
-(function (ETError) {
-    // React-native-ExecuTorch errors
-    ETError[ETError["UndefinedError"] = 101] = "UndefinedError";
-    ETError[ETError["ModuleNotLoaded"] = 102] = "ModuleNotLoaded";
-    ETError[ETError["FileWriteFailed"] = 103] = "FileWriteFailed";
-    ETError[ETError["ModelGenerating"] = 104] = "ModelGenerating";
-    ETError[ETError["LanguageNotSupported"] = 105] = "LanguageNotSupported";
-    ETError[ETError["InvalidModelSource"] = 255] = "InvalidModelSource";
-    //SpeechToText errors
-    ETError[ETError["MultilingualConfiguration"] = 160] = "MultilingualConfiguration";
-    ETError[ETError["MissingDataChunk"] = 161] = "MissingDataChunk";
-    ETError[ETError["StreamingNotStarted"] = 162] = "StreamingNotStarted";
-    // ExecuTorch mapped errors
-    // Based on: https://github.com/pytorch/executorch/blob/main/runtime/core/error.h
-    // System errors
-    ETError[ETError["Ok"] = 0] = "Ok";
-    ETError[ETError["Internal"] = 1] = "Internal";
-    ETError[ETError["InvalidState"] = 2] = "InvalidState";
-    ETError[ETError["EndOfMethod"] = 3] = "EndOfMethod";
-    // Logical errors
-    ETError[ETError["NotSupported"] = 16] = "NotSupported";
-    ETError[ETError["NotImplemented"] = 17] = "NotImplemented";
-    ETError[ETError["InvalidArgument"] = 18] = "InvalidArgument";
-    ETError[ETError["InvalidType"] = 19] = "InvalidType";
-    ETError[ETError["OperatorMissing"] = 20] = "OperatorMissing";
-    // Resource errors
-    ETError[ETError["NotFound"] = 32] = "NotFound";
-    ETError[ETError["MemoryAllocationFailed"] = 33] = "MemoryAllocationFailed";
-    ETError[ETError["AccessFailed"] = 34] = "AccessFailed";
-    ETError[ETError["InvalidProgram"] = 35] = "InvalidProgram";
-    ETError[ETError["InvalidExternalData"] = 36] = "InvalidExternalData";
-    ETError[ETError["OutOfResources"] = 37] = "OutOfResources";
-    // Delegate errors
-    ETError[ETError["DelegateInvalidCompatibility"] = 48] = "DelegateInvalidCompatibility";
-    ETError[ETError["DelegateMemoryAllocationFailed"] = 49] = "DelegateMemoryAllocationFailed";
-    ETError[ETError["DelegateInvalidHandle"] = 50] = "DelegateInvalidHandle";
-})(ETError || (ETError = {}));
-export const getError = (e) => {
-    if (typeof e === 'number') {
-        if (e in ETError)
-            return ETError[e];
-        return ETError[ETError.UndefinedError];
-    }
-    // try to extract number from message (can contain false positives)
-    const error = e;
-    const errorCode = parseInt(error.message, 10);
-    const message = Number.isNaN(errorCode)
-        ? error.message
-        : ' ' + error.message.slice(`${errorCode}`.length).trimStart();
-    const ETErrorMessage = (errorCode in ETError ? ETError[errorCode] : ETError[ETError.UndefinedError]);
-    return ETErrorMessage + message;
-};

package/lib/ThreadPool.d.ts

@@ -1,10 +0,0 @@
-declare class NativeThreadPool {
-    constructor();
-    executeNative(functionName: any, args: any): Promise<any>;
-    runInference(prompt: any, options?: {}): Promise<any>;
-    processImage(imagePath: any, options?: {}): Promise<any>;
-    heavyComputation(data: any): Promise<any>;
-    getStats(): any;
-}
-declare const _default: NativeThreadPool;
-export default _default;

package/lib/ThreadPool.js

@@ -1,28 +0,0 @@
-// ThreadPool.js
-class NativeThreadPool {
-    constructor() {
-        if (!global.NativeThreadPool) {
-            throw new Error('NativeThreadPool not installed');
-        }
-        this.pool = global.NativeThreadPool;
-    }
-    // Execute any native function
-    async executeNative(functionName, args) {
-        return await this.pool.executeNative(functionName, JSON.stringify(args));
-    }
-    // Specific methods for common tasks
-    async runInference(prompt, options = {}) {
-        return await this.executeNative('runInference', { prompt, ...options });
-    }
-    async processImage(imagePath, options = {}) {
-        return await this.executeNative('processImage', { imagePath, ...options });
-    }
-    async heavyComputation(data) {
-        return await this.executeNative('heavyComputation', data);
-    }
-    // Get thread pool statistics
-    getStats() {
-        return this.pool.getStats();
-    }
-}
-export default new NativeThreadPool();

package/lib/common/Logger.d.ts

@@ -1,8 +0,0 @@
-export declare class Logger {
-    private static readonly PREFIX;
-    static log(...data: any[]): void;
-    static debug(...data: any[]): void;
-    static info(...data: any[]): void;
-    static warn(...data: any[]): void;
-    static error(...data: any[]): void;
-}

package/lib/common/Logger.js

@@ -1,19 +0,0 @@
-/* eslint-disable no-console */
-export class Logger {
-    static PREFIX = '[React Native ExecuTorch]';
-    static log(...data) {
-        console.log(Logger.PREFIX, ...data);
-    }
-    static debug(...data) {
-        console.debug(Logger.PREFIX, ...data);
-    }
-    static info(...data) {
-        console.info(Logger.PREFIX, ...data);
-    }
-    static warn(...data) {
-        console.warn(Logger.PREFIX, ...data);
-    }
-    static error(...data) {
-        console.error(Logger.PREFIX, ...data);
-    }
-}

package/lib/constants/directories.js

@@ -1,2 +0,0 @@
-import { documentDirectory } from 'expo-file-system';
-export const RNEDirectory = `${documentDirectory}react-native-executorch/`;

package/lib/constants/llmDefaults.d.ts

@@ -1,6 +0,0 @@
-import { ChatConfig, Message } from '../types/llm';
-export declare const DEFAULT_SYSTEM_PROMPT = "You are a knowledgeable, efficient, and direct AI assistant. Provide concise answers, focusing on the key information needed. Offer suggestions tactfully when appropriate to improve outcomes. Engage in productive collaboration with the user. Don't return too much text.";
-export declare const DEFAULT_STRUCTURED_OUTPUT_PROMPT: (structuredOutputSchema: string) => string;
-export declare const DEFAULT_MESSAGE_HISTORY: Message[];
-export declare const DEFAULT_CONTEXT_WINDOW_LENGTH = 5;
-export declare const DEFAULT_CHAT_CONFIG: ChatConfig;

package/lib/constants/llmDefaults.js

@@ -1,16 +0,0 @@
-export const DEFAULT_SYSTEM_PROMPT = "You are a knowledgeable, efficient, and direct AI assistant. Provide concise answers, focusing on the key information needed. Offer suggestions tactfully when appropriate to improve outcomes. Engage in productive collaboration with the user. Don't return too much text.";
-export const DEFAULT_STRUCTURED_OUTPUT_PROMPT = (structuredOutputSchema) => `The output should be formatted as a JSON instance that conforms to the JSON schema below.
-
-As an example, for the schema {"properties": {"foo": {"title": "Foo", "description": "a list of strings", "type": "array", "items": {"type": "string"}}}, "required": ["foo"]}
-the object {"foo": ["bar", "baz"]} is a well-formatted instance of the schema. The object {"properties": {"foo": ["bar", "baz"]}} is not well-formatted.
-
-Here is the output schema:
-${structuredOutputSchema}
-`;
-export const DEFAULT_MESSAGE_HISTORY = [];
-export const DEFAULT_CONTEXT_WINDOW_LENGTH = 5;
-export const DEFAULT_CHAT_CONFIG = {
-    systemPrompt: DEFAULT_SYSTEM_PROMPT,
-    initialMessageHistory: DEFAULT_MESSAGE_HISTORY,
-    contextWindowLength: DEFAULT_CONTEXT_WINDOW_LENGTH,
-};

package/lib/constants/modelUrls.d.ts

@@ -1,223 +0,0 @@
-import { AvailableModels } from '../types/stt';
-export declare const LLAMA3_2_3B: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const LLAMA3_2_3B_QLORA: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const LLAMA3_2_3B_SPINQUANT: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const LLAMA3_2_1B: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const LLAMA3_2_1B_QLORA: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const LLAMA3_2_1B_SPINQUANT: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const QWEN3_0_6B: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const QWEN3_0_6B_QUANTIZED: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const QWEN3_1_7B: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const QWEN3_1_7B_QUANTIZED: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const QWEN3_4B: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const QWEN3_4B_QUANTIZED: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const HAMMER2_1_0_5B: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const HAMMER2_1_0_5B_QUANTIZED: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const HAMMER2_1_1_5B: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const HAMMER2_1_1_5B_QUANTIZED: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const HAMMER2_1_3B: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const HAMMER2_1_3B_QUANTIZED: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const SMOLLM2_1_135M: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const SMOLLM2_1_135M_QUANTIZED: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const SMOLLM2_1_360M: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const SMOLLM2_1_360M_QUANTIZED: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const SMOLLM2_1_1_7B: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const SMOLLM2_1_1_7B_QUANTIZED: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const QWEN2_5_0_5B: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const QWEN2_5_0_5B_QUANTIZED: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const QWEN2_5_1_5B: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const QWEN2_5_1_5B_QUANTIZED: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const QWEN2_5_3B: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const QWEN2_5_3B_QUANTIZED: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const PHI_4_MINI_4B: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const PHI_4_MINI_4B_QUANTIZED: {
-    modelSource: string;
-    tokenizerSource: string;
-    tokenizerConfigSource: string;
-};
-export declare const EFFICIENTNET_V2_S: {
-    modelSource: string;
-};
-export declare const SSDLITE_320_MOBILENET_V3_LARGE: {
-    modelSource: string;
-};
-export declare const STYLE_TRANSFER_CANDY: {
-    modelSource: string;
-};
-export declare const STYLE_TRANSFER_MOSAIC: {
-    modelSource: string;
-};
-export declare const STYLE_TRANSFER_RAIN_PRINCESS: {
-    modelSource: string;
-};
-export declare const STYLE_TRANSFER_UDNIE: {
-    modelSource: string;
-};
-export declare const MOONSHINE_TINY: {
-    modelName: AvailableModels;
-    decoderSource: string;
-    encoderSource: string;
-    tokenizerSource: string;
-};
-export declare const WHISPER_TINY: {
-    modelName: AvailableModels;
-    decoderSource: string;
-    encoderSource: string;
-    tokenizerSource: string;
-};
-export declare const WHISPER_TINY_MULTILINGUAL: {
-    modelName: AvailableModels;
-    decoderSource: string;
-    encoderSource: string;
-    tokenizerSource: string;
-};
-export declare const DEEPLAB_V3_RESNET50: {
-    modelSource: string;
-};
-export declare const CLIP_VIT_BASE_PATCH32_IMAGE: {
-    modelSource: string;
-};
-export declare const ALL_MINILM_L6_V2: {
-    modelSource: string;
-    tokenizerSource: string;
-};
-export declare const ALL_MPNET_BASE_V2: {
-    modelSource: string;
-    tokenizerSource: string;
-};
-export declare const MULTI_QA_MINILM_L6_COS_V1: {
-    modelSource: string;
-    tokenizerSource: string;
-};
-export declare const MULTI_QA_MPNET_BASE_DOT_V1: {
-    modelSource: string;
-    tokenizerSource: string;
-};
-export declare const CLIP_VIT_BASE_PATCH32_TEXT: {
-    modelSource: string;
-    tokenizerSource: string;
-};