mvt Releases

Focuses on engine installation reliability, dictation improvements, and UI refinements.

Sep 10, 2026

YuE2 introduces frontier song quality and editable symbolic plans, enabling users to create songs from lyrics, generate zero-shot covers from transcribed melodies, and revise compositions via agents.

Sep 9, 2026

Improves dashboard performance and configuration error reporting while fixing installation checksums.

Sep 8, 2026

This release focuses on performance optimizations, including caching invariant recording and facecam masks, adding native export benchmarks, and improving long-run validation. It also includes documentation updates for zoom effects and HUD actions.

Sep 8, 2026

Adds a context menu to the HUD for quitting the application and optimizes static video export performance by caching backgrounds.

Sep 7, 2026

A major update replacing the legacy iOSbackup library with iphone_backup_decrypt, introducing a new SysdiagnoseInfo module, and improving CI/CD coverage.

Sep 7, 2026

Introduces adaptive automatic zoom with protected camera paths and improves the zoom inspector UI by adding a delete-all menu and fixing changelog rendering.

Sep 6, 2026

This update addresses stability and security issues while introducing workflow management tools and improved support for Krea-2 LoRAs.

Sep 6, 2026

Resolves dashboard installation failures by fixing version parsing and ensuring Python 3.11+ compatibility.

Sep 5, 2026

Introduces traffic history tracking to the dashboard and resolves WEB connection binding issues.

Sep 5, 2026

Embeds Computer Use in the binary, adds cloud dispatch capabilities, introduces Alibaba Model Studio and Concentrate, and enables anonymous usage counting by default.

Sep 5, 2026

Minimal release notes: "## Commits since v0.2.46 - de8bafb Merge pull request #1080 from imbhargav5/codex/native-shortcut-recorder - 7729870 fe"

Sep 3, 2026

Minimal release notes: "## Commits since v0.2.45 - 4ff6060 Merge pull request #1076 from imbhargav5/codex/wider-tabbed-settings-dialog - c1253a"

Sep 3, 2026

A major release introducing an extensible CLI, a plugin system, Docker ARM64 support, and the addition of a WhatsApp contacts module.

Sep 1, 2026

A major feature release introducing video generation, support for several new text-to-image models, FP8 quantization for VRAM savings, and multi-GPU capabilities.

Aug 25, 2026

Tightens the long-running agent loop, makes workflow failures visible, and adds an experimental vision-capable DeepSeek route alongside portable config features.

Aug 23, 2026

This is the first versioned release of NeMo-Speech.cpp, providing a comprehensive toolkit for speech tasks including recognition, diarization, synthesis, and translation. It offers a unified CLI, OpenAI-compatible HTTP API, and support for multiple backends.

Aug 19, 2026

Ghidra 12.1.3 is a maintenance release with supplied notes that reference the official What's New, Change History, and Getting Started documents. The release text lists only the build checksum and does not describe individual changes. No breaking changes or new features are explicitly documented.

Aug 18, 2026

This update enhances the Custom Dictionary with advanced formatting options and fixes critical issues regarding AI response handling and microphone reliability in clamshell mode.

Aug 18, 2026

The release introduces offline speaker labeling capabilities for file transcription and a new microphone priority management system, alongside fixes for tool call streaming and application focus.

Aug 11, 2026