Skip to content

GCT v4.5.2 - Gemini 3.8 & 3.7 Flash Support & Smart Demo Audio

Choose a tag to compare

View all tags
@tomkam1702 tomkam1702 released this 08 Sep 21:11
· 2 commits to main since this release
v4.5.2
78d18f2

Release Date: 8 September 2026

Version 4.5.2 adds support for Google's newest frontier models (Gemini 3.8 Flash and Gemini 3.7 Flash) for translation with ultra-fast low-latency reasoning configuration, and enhances the Free trial TTS experience by dynamically matching announcement languages to the active mode.

What's New in Version 4.5.2

Frontier Gemini 3.8 Flash & Gemini 3.7 Flash Support

  • Gemini 3.8 Flash & Gemini 3.7 Flash for translation: Added full support for Google's latest models (gemini-3.8-flash and gemini-3.7-flash).
  • Optimised low-latency reasoning (thinking_level="low"): Fully resolves the 400 INVALID_ARGUMENT error caused by Google's removal of the MINIMAL thinking level in 3.7+ models. The models now respond swiftly with minimal reasoning delay, perfectly tailored for real-time game and movie subtitle translation.
  • Streamlined API payload: Deprecated legacy sampling parameters (temperature, top_p, top_k, candidate_count) are automatically omitted for Gemini 3.7+ models in accordance with official Google API specifications.

Context-Aware Voice-Over Demo Announcements

  • Smart language matching in Free trial: In the TTS Free trial tier (3-minute trial on every app launch), demo start and expiration announcements now dynamically match the active mode:
    • Spoken in the native source language when using Read Original Subtitles (No Translation).
    • Spoken in the target language when using translation modes (Voice-Over Only or Voice-Over and Subtitles).

Smart Full & Lite Update Packages

  • If Supertonic TTS neural models are already installed on your PC, the app automatically downloads the smaller Lite installer (~73 MB instead of ~428 MB) and seamlessly preserves existing models with automated backup and integrity verification.

📥 Downloads & Requirements

  • 🪄 Full Installer (with TTS models): GameChangingTranslator_v4.5.2_Full.exe (~428 MB)
  • ⚡ Lite Installer (without TTS models - for existing users): GameChangingTranslator_v4.5.2_Lite.exe (~73 MB)
  • 💻 System Requirements:
    • Windows 10 / 11 (64-bit)
    • Processor with AVX2 support (standard on CPUs from 2016 onwards)
    • 16 GB RAM recommended for offline Neural Voice-Over
    • Valid API key from Google (AI Studio) or DeepInfra
    • No dedicated GPU/CUDA required

📖 Resources

Game-Changing Translator v4.5.2 – Next-generation real-time AI screen translation & neural voice-over.

Assets 4
Loading