Skip to content
Submit a ticket
  • There are no suggestions because the search field is empty.

Connecting Model Match to Perplexity

Add Model Match as a custom connector in Perplexity to search Loan Officers, Real Estate Agents, Companies, lenders, and live market data

In this article:

  1. What the Model Match connector does
  2. The Model Match connector (MCP server) URL
  3. Adding the Model Match connector in Perplexity
  4. Enterprise accounts and admin permissions
  5. Confirming the connection works
  6. Frequently asked questions

What the connector does

The Model Match connector is a remote MCP server that gives Perplexity secure access to your Model Match workspace through the Model Context Protocol. Once it's connected, Perplexity can search and act on your data without you leaving the thread.

What it reaches:

  • Loan Officers and Real Estate Agents: search by market, production, and tenure, and pull contact details.
  • Companies and Branches: roll up volume, look up offices, and trace where an originator sends business.
  • Lenders and live market data: the same production and market activity you see in Market Insights.
  • Your own records: your CRM lists, notes, and pipeline — Perplexity can read them and update them.

Every tool call runs against your own Model Match sign-in, so Perplexity sees exactly what you see in the product — nothing more, and nothing that belongs to a teammate.

The Model Match connector (MCP server) URL

One address does the whole setup:

https://perplexity.modelmatch.com/mcp

Remote connector URLs must use HTTPS. This one already does — paste it exactly as written.

Adding the Model Match connector in Perplexity

⭐ Custom connectors are available on the Perplexity Pro, Max, and Enterprise plans. Connecting also requires an active Model Match account.

Let's add it from your Perplexity account settings; it takes about a minute, and there's no client ID or secret to hunt down.

  1. Open Settings → Connectors and, under Access, click + Custom connector. Choose Remote.
  2. Name it Model Match and paste the server URL above.
  3. Open Advanced settings and set Transport to Streamable HTTP.
  4. Set Authentication to OAuth. Model Match supports OAuth discovery and dynamic client registration, so there's no client ID or secret to paste — leave those fields empty.
  5. Tick the acknowledgement box and click Add, then sign in to Model Match when the authorization prompt appears.

💡 The acknowledgement box is Perplexity's standard warning for any custom connector, not a Model Match permission grant. Ticking it doesn't hand Perplexity anything beyond the workspace access your own login already carries.

Once you've signed in, the connector is live on your account and every tool call it makes respects your existing workspace roles and permissions.

Enterprise accounts and admin permissions

On an Enterprise workspace, an admin has to allow custom connectors before anyone can add one. In Enterprise settings, open Connectors and turn on Allow members to add custom connectors — it's off by default, so on a locked-down workspace the + Custom connector button won't get you anywhere until that's switched on.

An admin can instead add Model Match once for the whole organization. One catch worth knowing up front: an org-wide connector stays invisible to members until its creator shares it from the Permissions screen. If your team added Model Match and nobody can find it, that share step is almost always what's missing.

🧠 Each member still signs in with their own Model Match account, so access follows the person, not the connector. An org-wide connector distributes the setup — it never pools anyone's data.

Confirming the connection works

Start a new thread and ask for something only Model Match can answer:

"Search Model Match for the top loan officers in Austin, TX this year."

You should get named originators with production figures, not a web summary. If the thread returns a sign-in prompt instead of results, the authorization didn't finish — reopen Settings → Connectors and complete it.

Naming Model Match in the prompt is the reliable way to point Perplexity at the connector instead of the open web.

Frequently asked questions

Do I need a paid Perplexity plan for this? Yes — custom connectors are available on Pro, Max, and Enterprise. You'll also need an active Model Match account, but the connector authenticates as you, so there's no separate seat or license to buy on our side.

Is my data safe? What can Perplexity actually see? Perplexity connects using your own Model Match sign-in, and every tool call respects your existing workspace roles and permissions. It reaches the same records you'd see in Model Match — nothing a teammate has kept to themselves, and nothing outside your workspace.

Where do I paste the client ID and secret? There isn't one to paste. Model Match supports OAuth discovery and dynamic client registration, so setting Authentication to OAuth is all Perplexity needs — it negotiates credentials with Model Match on its own and hands you a sign-in prompt instead.

My admin added Model Match for the whole company, so why can't I see it? An org-wide connector stays invisible to members until its creator shares it from the Permissions screen in Enterprise settings. Ask whoever created it to share it, then sign in with your own Model Match account when it appears in your Connectors list.

Why isn't Perplexity using Model Match when I ask about a loan officer? Perplexity picks its sources thread by thread, and a bare question about a name reads like a web search. Say the surface out loud — "Search Model Match for…" — and it routes to the connector.

Can I connect Model Match to Claude or ChatGPT at the same time? Yes. One Model Match account can back connectors in several assistants at once, each authenticating as you.

See Connecting Model Match to Claude (Web, Desktop & Claude Code) and Connecting Model Match to ChatGPT (for OpenAI Users).