v4.5.81 v4.13
Graphify-Labs/graphifySep 19, 2026 → Sep 20, 20261 day between releases
What's New
New release introduces video transcription features and AI provider support, differing significantly from the previous CI-focused release.
Breaking Changes
- Refactored settings window and menu architecture
- Qwen-TTS/Higgs local channels dropped 8-bit quantization
New Features
- Pre-packaged Windows 10/11 download available
- Automatic model download retry on failure
- New AI provider support (Doubao 2.0, SiliconFlow, Minimax, OpenRouter)
- Media merge pause button
- API Route provider support
Improvements
- Optimized Whisper sentence segmentation algorithm
- Expanded support for new AI providers
- Refactored settings window and menu architecture
- Qwen-TTS/Higgs local channels dropped 8-bit quantization for compatibility
- Improved model downloading algorithms
Bug Fixes
- Fixed other known bugs
Upgrade Notes
For large version jumps, use the full package instead of the patch to avoid errors. Ensure dependencies like ffmpeg and libsndfile are installed correctly.
The Two Releases
v4.5.81
Sep 19, 2026· 1 month ago
Minimal release notes: "## What's Changed * ci(release): publish to PyPI without Docker by @carlotestor in https://github.com/ccxt/ccxt/pull/305"
Full release noteswindows-v4.13
Sep 20, 2026· 1 month ago
Released a pre-packaged Windows version (v4.13) of pyvideotrans, featuring improvements to model downloading algorithms, Whisper segmentation, and expanded support for new AI providers like Doubao 2.0.
Full release notes