Skip to content

Msty: using the VIDAI Control Plane as the backend

Msty's Remote Model Providers panel accepts custom OpenAI-compatible endpoints. Add one for the control plane and its models appear in Msty's model picker.

TL;DR

  1. Settings → Remote Model Providers → Add Provider. Pick the Open AI Compatible type.
  2. Fill in:
  3. Provider Name: VIDAI Control Plane
  4. API Endpoint: https://your-vidai-server.example.com/v1
  5. API Key: your VIDAI API key
  6. Under Models, add each model registered on the control plane that Msty should be able to pick (e.g. gpt-4o-mini, claude-sonnet-4.5).
  7. Save. Open a chat, choose your VIDAI model.

Prerequisites

  • Msty installed (macOS, Windows, or Linux desktop).
  • Control plane base URL and an API key from API Keys.
  • Models registered on the Models page.

Split Chats and Parallel Chat

Msty's Split Chat and Parallel Chat features run the same prompt against multiple models. Selecting your VIDAI model in one pane sends that pane's traffic through the control plane; the other pane can run a local model or a different remote provider — they don't interfere.

Attribution

One VIDAI API key per Msty install. For team deployments where several people share a Msty license, mint one key per user and have each user paste their own — Msty stores the provider config locally per install.

Verify it works

Pick your VIDAI model and send:

reply with exactly: ok

A row appears on Request Logs.

If something's off

Raise an issue at github.com/vidaiUK/vidai-quickstart/issues with the provider config (masking the key). We'll get it sorted.

Where to go next