Home Original page

GitHub - kikyo2006/XUnity.AutoPollinationTranslator: A translation endpoint for XUnity.AutoTranslator using Pollinations AI.

Built with Pollinations

A translation endpoint for XUnity.AutoTranslator that uses Pollinations AI to provide free, uncensored, and high-quality translations for games.

This project is a fork of XUnity.AutoChatGptTranslator optimized for the Pollination AI service.

Features

  • Free & Uncensored: Uses Pollinations AI which does not have restrictive content filters, making it ideal for various game contexts.
  • High Performance: Optimized JSON output handling for stable and clean translations.
  • Parallel Requests: Supports efficient translation batches.
  • Customizable: Full control over system prompts, models, and temperature.

Installation Instructions

  1. Build the Assembly: Compile the project or obtain the XUnity.AutoPollinationTranslator.dll.
  2. Install AutoTranslator: Ensure you have XUnity.AutoTranslator installed in your game.
  3. Deploy the Plugin: Place XUnity.AutoPollinationTranslator.dll into the <GameDir>/ManagedData/Translators directory (where other translator DLLs are located).
  4. Configure the Service: Update your AutoTranslatorConfig.ini file located in <GameDir>/AutoTranslator.

Configuration (AutoTranslatorConfig.ini)

Add or update the following sections in your configuration file:

[Service]
Endpoint=PollinationTranslate

[Pollination]
; Base URL for the service. Default is text.pollinations.ai
BaseUrl=https://gen.pollinations.ai/v1/chat/completions

; Model to use (options: openai, grok, mistral, p1, etc.)
Model=openai

; Your custom translation prompt
Prompt=Translate the following text to English, maintaining the original tone and context.

; Delay between translation requests in seconds (default is 1.0)
TranslateDelay=1.0

; Random seed for deterministic results (-1 for random)
Seed=-1

; Creativity level (0.0 = literal, 1.0 = balanced, 2.0 = creative)
Temperature=1.0

; Max retries for failed or truncated translations
MaxRetries=3

; Maximum tokens per response
MaxTokens=2000

; Enable debug logging in the Xua console
Debug=false

Credits

  • Original Author: joshfreitas1984 for the initial ChatGPT translator implementation.
  • Service Provider: Pollinations AI for the underlying AI infrastructure.