Bilargo Translator — AI & Model Notice

Effective date: September 9, 2026

This notice supplements the Privacy Policy and Terms of Use.

1. Where translation runs

Select an on-device engine, a server on this Mac, a server on another LAN computer, or an API provider. A LAN server on another computer receives text off your Mac. Bilargo operates no inference server and receives none of these requests. Server software and provider policies determine their own logging and onward processing.

2. Model catalog

Apple Translate

Apple’s Translation framework processes text on this Mac. Available pairs depend on macOS and may require Apple language-pack downloads before offline use.

TranslateGemma and Gemma 4 MLX

The catalog includes TranslateGemma 4B/12B and Gemma 4 MLX translation models. Download a model before local inference. The language picker lists 49 languages; support and quality depend on the selected engine and language pair. The app interface is available in English and Polish, with German, French, and Spanish development drafts labelled “Deutsch (Entwurf)”, “Français (brouillon)”, and “Español (borrador)” in Interface language. All three draft catalogs have complete key coverage and automated checks; native-speaker, visual, VoiceOver, and minimum-window reviews remain NOT RUN. System can select a draft automatically when it matches macOS language preferences. Model names and descriptions may remain in their source language. Interface languages are separate from model translation pairs.

TranslateGemma is a translation-specialist model. Glossary handling depends on model capabilities: check the resulting terminology. Gemma 4 is also available for translation.

On-device text recognition (OCR)

Vision and bundled Tesseract resources read images and scanned PDF pages on this Mac. Recognition accuracy and languages depend on available resources. Extracted text is translated by the selected engine, so it leaves this Mac if you choose another LAN computer or an API provider. PDF results are translated text, not reconstructed PDF layout.

Servers and API providers

You can configure an OpenAI-compatible server, such as Ollama or LM Studio, on this Mac or another computer. You can also configure an API provider using your own key. API use is charged by the provider separately from the app purchase. The provider’s terms and the hosted model’s license apply. See Privacy Policy, sections 4a–4b.

3. Model sources and licenses

Model family License / Terms Links
TranslateGemma Gemma Terms of Use + Prohibited Use Policy Gemma Terms · Prohibited Use Policy
MLX community conversions Original model license applies; conversion tooling under MIT mlx-community on Hugging Face
Tesseract OCR + Leptonica (bundled OCR fallback) Apache License 2.0 Tesseract · Leptonica

Users who download or use TranslateGemma models must comply with the Gemma Terms of Use and the Gemma Prohibited Use Policy.

4. Privacy: how models interact with your data

5. Limitations and disclaimers

6. Model availability

Model availability may change for legal, licensing, security, or technical reasons. We may add, update, or remove support for any model at our discretion, subject to your rights under applicable consumer law and the Terms of Use.

7. Contact

For questions about AI models or this notice: