Reference Vault v1.0.2

Reference Vault v1.0.2

This major update overhauls the app's architecture for massive datasets and introduces Universal GPU Acceleration. Reference Vault now dynamically scales from lightweight laptops to high-end studio workstations.

✨ New Features

• Tree-View Sidebar: The folder sidebar has been completely rebuilt. Nested subfolders now visually branch out under their parent folders, making it infinitely easier to navigate complex libraries.
• Live Auto Tagger Queue Tracker: Added a real-time status tracker to the bottom of the sidebar. You will now see exactly how many images the Tagger has left to tag
• Universal Hardware Support: The AI engine has been upgraded to use Microsoft DirectML. It now natively utilizes NVIDIA, AMD, and Intel graphics cards via DirectX 12.
• Empty Folder UX: Clicking a master folder that only contains subfolders will no longer show a confusing blank screen. A helpful prompt now directs you to click the subfolders in the sidebar.
• Automatic Update Checker: The app will now silently check GitHub on startup. If a new version of Reference Vault drops, you will get a popup with a direct download link!
• Smart Fallback: If you download the GPU version but your hardware doesn't support it, the app will automatically catch the error and seamlessly boot the CPU engine instead.

✨ New Features

• Interactive Tree Sidebar: The folder list is now a fully collapsible visual tree with native OS icons. Nested subfolders display perfectly, making it easy to navigate complex reference libraries.
• Manual Tag Editing: Disagree with the AI? Right-click any image in your grid to manually edit, delete, or add custom searchable tags.
• Safe Folder Renaming & Creation: Right-click any folder in the sidebar to securely rename it without breaking your AI tags. Right-clicking and selecting "+ New Folder" will now intelligently nest the new folder directly inside the parent.

🛠️ Bug Fixes & Optimizations

• The "PIL" Crash Fix: Fixed a critical bug where the AI engine would fail to read images on certain Windows machines due to a missing Pillow dependency in the compiled executable.
• Thread Throttling & UI Smoothness: The background AI thread has been re-engineered to take microscopic pauses between image processing. This ensures the main app window remains perfectly smooth and responsive, even when chewing through thousands of images.
• Local Crash Logging: If the AI engine hits a fatal error , It will automatically generate a Vault_Crash_Log.txt file in the installation folder for easy debugging.
• Zero-Lag Interface: Engineered a new "Signal Throttling" and "SQLite Batch Transaction" system. You can now drop thousands of images into the app, and the main window will remain buttery smooth while the AI processes them in the background.

📥 Which version should I download?

• Standard CPU Only Edition (~52MB): Perfect for laptops or older PCs.
• GPU Edition (~60MB): Uses your graphics card with DirectX12 to tag massive libraries at blistering speeds.

📥 How to Install

1. Download the ReferenceVault_v1.0.2_Setup.exe file for CPU only or 'ReferenceVault_v1.0.2_GPUSetup.exe' file for GPU support.
2. Run the installer (it will safely overwrite your previous installation while keeping your saved databases intact).
3. Note: If Windows SmartScreen flags the app because it is an unsigned indie executable, click More info -> Run anyway.