pixlvault 0.8.7

Self-hosted picture library with AI tagging, semantic search, and a clean web UI

PixlVault

PixlVault is a local picture library server for organizing, filtering, and reviewing large image collections.

It provides:

• A browser-based interface
• Fast metadata and tag filtering
• Smart score sorting
• Character and set organization
• Local storage of your library data

PixlVault runs on your machine and serves the UI at a local web address.

Install PixlVault

Choose one installation method.

Option 1: Windows installer

Use this if you want the easiest setup on Windows.

1. Go to the GitHub Releases page for this repository.
2. Download the latest Windows installer .exe.
3. Run the installer.
4. Start PixlVault Server from the Start Menu shortcut.
5. Open your browser to http://localhost:9537.

Windows SmartScreen warning: Because the installer is not yet signed with a paid code-signing certificate, Windows SmartScreen may show a red "Windows protected your PC" dialog when you run it. This is expected. Click More info and then Run anyway to proceed with the installation.

Option 2: Install from PyPI

Use this if you already have Python and want a pip install.

Requirements:

• Python 3.10 or newer

Install:

pip install pixlvault

Run:

pixlvault-server

Then open:

http://localhost:9537

Option 3: Clone and run manually

Use this if you want to run from source.

Requirements:

• Python 3.10 or newer
• Node.js 20 or newer
• npm

Steps:

git clone https://github.com/Pixelurgy/pixlvault.git
cd pixlvault

python -m venv .venv
# Windows:
.venv\Scripts\activate
# macOS/Linux:
# source .venv/bin/activate

pip install --upgrade pip
pip install -e .

cd frontend
npm ci
npm run build
cd ..

pixlvault-server

Then open:

http://localhost:9537

First run and data location

On first run, PixlVault creates a user config directory and stores:

• Server config
• Database
• Imported media files

Model downloads: On first startup, PixlVault automatically downloads the AI models required for tagging, captioning, and quality scoring. This includes several hundred MB of model weights. Downloads happen in the background and are stored in the platform user data directory:

OS	Path
Linux	~/.local/share/pixlvault/downloaded_models/
macOS	~/Library/Application Support/pixlvault/downloaded_models/
Windows	%LOCALAPPDATA%\pixlvault\downloaded_models\

An internet connection is required the first time the server starts. Subsequent starts use the cached models.

If you need to use a custom config path:

python -m pixlvault.app --server-config "C:\path\to\server-config.json"

Server configuration

On first run, PixlVault generates a server-config.json file in the user config directory:

• Linux / macOS: ~/.config/pixlvault/server-config.json
• Windows: %APPDATA%\pixlvault\server-config.json

You can also supply a custom path with --server-config <path>.

Edit the file and restart the server to apply changes.

Network and port

Key	Default	Description
host	"localhost"	Address the server binds to. Change to "0.0.0.0" to expose the server on the local network.
port	9537	TCP port the server listens on.
cors_origins	[]	Extra origins allowed to make credentialed cross-origin requests. localhost, 127.0.0.1, and the server's own LAN IP are always permitted on any port.

At startup the server detects its own LAN IP and automatically allows it on any port. This means the Vite dev server works over LAN (http://192.168.1.5:5173 → http://192.168.1.5:9537) without any extra configuration, as long as network access is enabled via host.

Use cors_origins only if you need to allow origins on a different machine entirely.

SSL / HTTPS

Key	Default	Description
require_ssl	false	Enable HTTPS. When true, the server will use the key and certificate below.
ssl_keyfile	<config_dir>/ssl/key.pem	Path to the SSL private key file.
ssl_certfile	<config_dir>/ssl/cert.pem	Path to the SSL certificate file.
cookie_samesite	"Lax"	SameSite attribute for session cookies ("Lax", "Strict", or "None").
cookie_secure	false	Set the Secure flag on session cookies. Enable when serving over HTTPS.

Storage

Key	Default	Description
image_root	<config_dir>/images	Directory where imported media files are stored.
watch_folders	[]	List of folder entries to watch for new images and automatically import them. Each entry is an object with the fields below.

Each entry in watch_folders has the following fields:

Field	Type	Default	Description
folder	string	—	Absolute path to the directory to monitor (recursively).
delete_after_import	boolean	false	When true, source files are deleted from the watch folder after a successful import.

Example:

"watch_folders": [
  { "folder": "/home/user/downloads/photos", "delete_after_import": false },
  { "folder": "/mnt/camera", "delete_after_import": true }
]

Processing

Key	Default	Description
default_device	"cpu"	Device used for AI processing ("cpu" or "cuda").
generate_thumbnails_on_startup	true	Generate missing thumbnails when the server starts.

Logging

Key	Default	Description
log_level	"info"	Log verbosity ("debug", "info", "warning", "error").
log_file	<config_dir>/server.log	Path to the log file.

Example config

{
  "host": "localhost",
  "port": 9537,
  "log_level": "info",
  "require_ssl": false,
  "image_root": "/home/user/.config/pixlvault/images",
  "watch_folders": [
    { "folder": "/path/to/photos", "delete_after_import": false }
  ],
  "default_device": "cpu",
  "generate_thumbnails_on_startup": true
}

Updating PixlVault

PyPI install

pip install --upgrade pixlvault

Source install

Pull latest changes, rebuild frontend, and reinstall:

git pull
cd frontend
npm ci
npm run build
cd ..
pip install -e .

Troubleshooting

• If the page does not load, confirm the server process is running.
• If port 9537 is in use, set a different port in your server config file.
• If frontend assets are missing, rebuild frontend with npm run build and restart the server.

GPU startup fails (CUDAExecutionProvider unavailable)

If startup reports that ONNX CUDAExecutionProvider is unavailable, you likely have CPU-only ONNX Runtime installed.

Fix your environment:

pip uninstall -y onnxruntime
pip install onnxruntime-gpu

Verify providers:

python -c "import onnxruntime as ort; print(ort.get_available_providers())"

Expected output should include CUDAExecutionProvider.

If you prefer CPU mode, set "default_device": "cpu" in server-config.json.

Installing plugins

PixlVault supports built-in plugins and user-created plugins.

User plugin directory

Place your .py plugin files in the platform-specific user data directory. PixlVault logs the exact path on startup.

OS	Path
Linux	~/.local/share/pixlvault/image-plugins/user/
macOS	~/Library/Application Support/pixlvault/image-plugins/user/
Windows	%LOCALAPPDATA%\pixlvault\image-plugins\user\

Writing a plugin

Use the template from image-plugins/user/plugin_template.py in the source repository as a starting point:

1. Create a new .py file in your user plugin directory.
2. Subclass ImagePlugin, set a unique name and plugin_id, and implement run().
3. Restart PixlVault Server — plugins are loaded at startup.

plugin_template.py is ignored by plugin discovery and will not be loaded as a plugin.