ʻŌlelo Honua

🌺🌸🌼 Bloom Your Internationalization Workflow! 🌷🌻🌹

🌟 Meet Our Smarter, Faster AI Language Engine! 🌟

Ready to supercharge your i18n workflow? Our newly revamped AI language engine is here to make translations smarter, faster, and more intuitive. It delivers spot-on, culturally aware translations in record time, with built-in support for async and parallel processing. Whether you’re scaling up or just starting out, this engine grows with you—making multilingual app development effortless and efficient. Say hello to the future of i18n!

Simplify Your Internationalization Workflow with ʻŌlelo Honua

Struggling to keep up with the demands of managing translation files for your multilingual app? Tools like i18next or react-i18next often leave you stuck in manual workflows, relying on services like Google Translate or ChatGPT to fill the gaps. It’s tedious, time-consuming, and easy to make mistakes.

That’s where ʻŌlelo Honua comes in. This tool takes the hassle out of internationalization (i18n) by automating translation file creation and updates. Beyond just saving time, it ensures your translations are accurate, culturally appropriate, and free from common errors.

Why waste time on repetitive tasks when you can streamline your i18n process with a tool built to simplify localization? Let ʻŌlelo Honua handle the hard parts so you can focus on delivering great experiences for your global users.

Challenges in Multilingual App Development and How ʻŌlelo Honua Solves Them

Problem

Imagine building an app that speaks to the world. You start with one language, but as your app grows, so does the demand for multilingual support. Manually crafting and updating translation files for every new language? It’s a grind—tedious, error-prone, and a major time sink. Sound familiar?

With these models, ʻŌlelo Honua ensures you have access to the best tools for your internationalization needs, whether you’re working on a small project or scaling to global audiences.

Solution

Say goodbye to the headaches of managing translation files! ʻŌlelo Honua takes the grunt work out of internationalization by automating the creation and synchronization of translation files. With support for trusted providers like Google Translate, it ensures your translations are not only accurate but also always up-to-date. Save time, eliminate errors, and focus on what truly matters—building a world-class multilingual app.

Here’s a simple illustration:

1. Without ʻŌlelo Honua: You manually create and update translation files for each language. Commence repetitive copy/paste from Google, ChatGPT, etc… 😩
2. With ʻŌlelo Honua: The library automatically generates and syncs translation files using your preferred translation provider. 🎩✨

This automation makes it easier to manage internationalization in your application, allowing you to focus on other stuff!

Technical Deep Dive

For more details about my approach, vision, and methodology, refer to About ʻŌlelo Honua.

Installation

To install the library, use npm or yarn:

npm install olelo-honua

or

yarn add olelo-honua

Usage

Here’s how you use the library:

const { OleloHonua, DeepSeekProvider } = require("olelo-honua");
// Create a new instance
const dakine = new OleloHonua(
  {
    provider: {
      platform: OleloHonua.Providers.OpenRouter,
      credentials: {
        apiKey: process.env.OPEN_ROUTER_API_KEY, // Open Router API Key from .env
      },
      modelId: OleloHonua.OpenRouterModels.DEEPSEEK.DEEPSEEK_V3_FREE, // Model ID
    },
    primeLanguage: "en", // Default language
    excludeLanguage: ["haw"], // Additional languages
    debug: true, // Enable debug
  },
);
// Runs or re-runs i18n translations
dakine.hanaHou(); // or use alias dakine.createLocaleFiles()

Configuration

I highly recommend setting up an OpenRouter API Key for use with various models.

How to set up OpenRouter API Key

There are also several other providers including Google Translate (aka Cloud Translation) and OpenAI (aka ChatGPT).

OpenAI API Quickstart

Google Cloud Translation Quickstart

For enhanced security and ease of configuration, you should store your API key in a .gitignore‘d .env file.

How to set up .gitignore for your project

Simply create a .env file in your project root and add the following line:

OPENROUTER_API_KEY=<your_openrouter_api_key>

Then, use a library like dotenv to load the key into your application:

require("dotenv").config();
const apiKey = process.env.OPENROUTER_API_KEY;

How to set up dotenv for your project

While using .gitignore, dotenv & .env is not mandatory for local development, it’s best engineering practice for safeguarding sensitive information. This approach keeps your credentials out of your codebase, simplifies environment management, and enhances security across different deployment setups.

You need to provide a configuration object and a translation provider. For instance, you can use the DeepSeekProvider as demonstrated in the usage example. I highly recommend DeepSeek & OpenRouter as these are free and extremely stable. Happy coding!

Configuration Example

Here is an example of full configuration options:

const config: Config = {
  primeLanguage: "en", // The primary language to use
  provider: {
    platform: "OpenRouter", // Supported platforms: 'OpenRouter' | 'OpenAI' | 'GoogleTranslate' | 'Custom'
    credentials: {
      apiKey: "<your_openrouter_api_key>", // Required for OpenRouter and OpenAI
    },
    modelId: "deepseek/deepseek-chat-v3-0324:free", // Model ID for OpenRouter and OpenAI
  },
  retries: {
    mainLoop: 3, // Retry limit for the main loop
    critiqueLoop: 2, // Retry limit for critique loop
    repairLoop: 1, // Retry limit for repair loop
  },
  debug: false, // Enable or disable debug mode
  includeLanguage: ["haw", "ar", "es", "fr"], // Languages to include, or alteratively...
  // excludeLanguage: ["haw", "ar", "es", "fr", "amh", "tg", "uz", "km", "lo", "my", "ne", "si", "ky", "mn", "ps", "sd", "tk", "ug", "xh", "zu"],
  maxChunkRequests: 4, // Maximum number of chunk requests
  additionalConfig: {
    critique: true, // Enable AI critique for translations
    saveCritique: false, // Optionally save critiques for further analysis
    repair: false, // Enable or disable automatic repair of translations
    multiLanguageAgreementThreshold: 0.8, // Set threshold for multi-language agreement
  },
};

Supported Providers

Olelo Honua integrates with OpenRouter, OpenAI/ChatGPT, Google Translate, and even local LLM providers.

Supported Models

ʻŌlelo Honua integrates with a variety of models to streamline your translation workflow. Below is a list of supported models categorized by provider:

OpenRouter Models

• DeepSeek
  • deepseek/deepseek-chat-v3-0324:free
  • deepseek/deepseek-r1:free
  • deepseek/deepseek-chat:free
• Google
  • google/gemini-2.5-pro-exp-03-25:free
  • google/gemma-3-1b-it:free
  • google/gemma-3-4b-it:free
  • google/gemma-3-12b-it:free
  • google/gemma-3-27b-it:free
  • google/gemini-2.0-flash-lite-preview-02-05:free
  • google/gemini-2.0-pro-exp-02-05:free
  • google/gemini-2.0-flash-thinking-exp:free
  • google/gemini-2.0-flash-thinking-exp-1219:free
  • google/gemini-2.0-flash-exp:free
  • google/learnlm-1.5-pro-experimental:free
  • google/gemma-2-9b-it:free
• Llama2
  • sophosympatheia/rogue-rose-103b-v0.2:free
  • gryphe/mythomax-l2-13b:free
• Llama3
  • deepseek/deepseek-r1-distill-llama-70b:free
  • meta-llama/llama-3.3-70b-instruct:free
  • nvidia/llama-3.1-nemotron-70b-instruct:free
  • meta-llama/llama-3.2-3b-instruct:free
  • meta-llama/llama-3.2-1b-instruct:free
  • meta-llama/llama-3.1-8b-instruct:free
  • meta-llama/llama-3-8b-instruct:free
• Qwen
  • qwen/qwq-32b:free
  • deepseek/deepseek-r1-distill-qwen-32b:free
  • deepseek/deepseek-r1-distill-qwen-14b:free
  • qwen/qwq-32b-preview:free
  • qwen/qwen-2.5-coder-32b-instruct:free
  • qwen/qwen-2.5-72b-instruct:free
  • qwen/qwen-2-7b-instruct:free
• Mistral
  • mistralai/mistral-7b-instruct:free

OpenAI Models

• OpenAI
  • openai/davinci
  • openai/curie
  • openai/babbage
  • openai/ada
  • openai/gpt-3.5-turbo
  • openai/gpt-4o

Note: Free models, such as OpenRouterProvider, come with a token limit. Ensure your usage stays within the allowed quota to avoid interruptions.

Select the provider/model that aligns with your application’s specific needs (and performance requirements) 🌟.

Caching

Translations are cached to keep things running smoothly and to cut down on API calls to translation providers. This is super handy if you’re using free or limited-tier services with strict quotas. By saving translations locally, you avoid hitting those limits and keep performance steady. If you ever need to clear the cache, just run this command:

rm -rf .translations_cache.json

Supported Languages

In addition to Hawaiian 🌺, ʻŌlelo Honua supports a wide range of languages:

Code of Conduct

We’re committed to fostering a welcoming and inclusive community. You can read our Code of Conduct here.

License

This project is licensed under the MIT License.