Platform

Platform overview How it works Authorization testing Evidence & reports Private scanning Integrations

Solutions

Security agencies Product teams Regulated industries Partner programme

Learn

Blog Knowledge hub Compare

Resources

Pricing Documentation Help centre FAQ Security & data What we haven’t proved

Company

About Contact Careers Sign in to the platform Start a $199 pilot →

Home / Help centre / Workspace settings

Workspace settings

How do I use a model that isn't in the dropdown?

Quick answer

Open Settings, choose AI Provider, then pick Custom… in the dropdown for the tier you want. Type the exact model id from your AI provider and click Save models. This lets you adopt a new model without waiting for an update.

Steps

Each of the three tiers has its own dropdown. You can set a custom id on one tier and leave the others alone.

STEPS

Good to know

The dropdown lists known models, and Custom… lets you type any model id. You are not tied to the built-in list. Enter the id exactly as your provider writes it, with no spaces.

The app does not show a list of valid ids. A wrong id is a typo that only a real run will catch, and by then the scan has already started and may have used up part of your allowance. The Verify button checks your credential, not the model name, so a passed check does not prove the id is right.

Your credential still applies to every tier. A custom id must be one that your credential is allowed to use.

Choose the tier with care. Deep analysis does the hardest reasoning, so changes there matter most. Read how to pick a model for each tier first if you have not seen the three dropdowns.

A premium model on a subscription token can be throttled part-way through a scan. Use an API key for those. Details are in the API key and OAuth token guide.

If a custom id stops working, return to the defaults with these steps.

Keep a note of the id you typed and the date. When a result changes after an edit, that note tells your team in seconds which setting moved, and it makes a rollback easy. It is a small habit with a large payoff. Because each tier reads its own dropdown, you can also try a new model on the recon tier alone while deep analysis stays on a model you already trust.

Full tutorial

AI provider settings: included allowance, your own credential and models

Step-by-step, with annotated screenshots.

Updated · Back to the help centre

Still stuck?

Tell us what you see on screen and an engineer will help you sort it out.