Purpose
Tolgee AI translator is the AI translation feature in the open-source Tolgee localization platform. Its goal is to complete application localization translation in less time and at lower cost, while improving final quality through human review.
Main capabilities
- Context-aware translation: Tolgee's context technology takes the context into account during translation, and according to official statements its results are better than ordinary machine translation. The integration can automatically extract context from the application, without needing to ask developers or designers for descriptions or screenshots.
- Supplementary context: Project descriptions, language-specific instructions, and translation key descriptions can be provided to help the AI understand the domain and tone, thereby generating more accurate and natural translations.
- AI translation + human review: First use AI translation, then have human translators review and polish it to a higher quality.
- Prompt customization: The default prompt fragments can be edited segment by segment, using variables to access key data, screenshots, and context, and custom prompts can be written through the editor.
- AI Playground: Before the formal translation project, test prompts with real translation data; you can switch between different LLM models, modify context, or define custom prompts, preview the results for the selected keys, and after tuning, save and set them as default.
- Connect your own AI model: You can connect any LLM provider, such as OpenAI, Anthropic, Google Gemini (including self-hosted solutions); you can configure multiple providers and use intelligent load balancing to avoid rate limits.
- Integration with AI coding assistants: Connect tools such as Claude Code, Cursor, and Codex through the MCP server, allowing AI to find translation keys, update translations, and directly generate machine translations in the editor.
How to use
Official materials show the following entry points:
- Use AI Playground to test prompts
- Customize prompts
- Configure LLM providers
- Explore the MCP server
Specific installation commands, code examples, and configuration details are not provided in this material.
Applicable scenarios
- Development teams that need to localize applications into multiple languages.
- Workflows that want to use AI to quickly generate a first draft before human translation, and then refine it manually.
- Developers who use AI coding assistants such as Claude Code, Cursor, and Codex and want to manage translations directly in the editor.
- Projects that need to self-host LLMs or connect to multiple model providers.
Limitations
- Specific pricing, costs, free quotas, and similar information are not provided in the material.
- Apart from the Spanish, German, and French mentioned in the examples, the material does not provide a complete list of supported languages.
- Specific integration steps, MCP installation methods, and code content are not provided in the material.provider projects.
Limitations
- Specific pricing, costs, free quotas, and similar information are not provided in the material.
- Apart from the Spanish, German, and French mentioned in the examples, the material does not provide a complete list of supported languages.
- Specific integration steps, MCP installation methods, and code content are not provided in the material.
