Vetrra Deterministic pipeline orchestration for local media libraries

INSTALLATION GUIDE

Install the desktop stack deliberately.

Vetrra is not a browser product with hidden infrastructure. The installation process should leave every dependency visible, callable, and easy to validate before the first real media job enters the queue.

1. Prepare the host

  • Use a Windows machine with stable access to your intake, staging, QC, and deploy paths.
  • Confirm the account running Vetrra can read source media and write final outputs.
  • Reserve enough free storage for transcode scratch space and remux staging.

2. Install the media tooling

  • Install FFmpeg and verify the binary is available on the system path.
  • Install MKVToolNix so stream inspection and mux policy are available to the pipeline.
  • Install SABnzbd if intake should originate from a downloader queue.
  • Install Ollama and pull the vision model you intend to use for poster parsing.

3. Validate command availability

ffmpeg -version
ffprobe -version
mkvmerge --version
ollama list

If any command fails here, stop and fix the local environment before troubleshooting the desktop application.

4. Confirm path policy before launch

  • Keep intake separate from deploy.
  • Keep QC output separate from final library placement.
  • Keep temporary encode paths on storage that can tolerate heavy write activity.

5. First launch checklist

  • Open Vetrra with one known-good title.
  • Confirm the queue appears in the operator surface.
  • Confirm the encode stage exposes the intended acceleration path.
  • Confirm subtitles, OCR, and deploy logs all render visible state instead of silent completion.

6. Beta 2 expectation

The installation path is still part of active Beta 2 hardening. Treat setup as an operator validation pass and document any dependency assumptions that differ from your environment.