Skip to main content
Each enrolled device is one seat. Change the type on the device. Owners, admins, and staff can make that change. Members can view it. A standard seat offers one model in the app: MSPilot Auto. That choice is Claude Sonnet 5. The name in the picker comes from branding and starts as MSPilot Auto. A power seat offers MSPilot Auto plus the models enabled for that client. Claude Desktop lists the Claude models. ChatGPT lists the GPT models. The current catalog is:
  • Claude Sonnet 5
  • Claude Opus 5.5
  • Claude Fable 5.1
  • GPT-5.6 Terra
  • GPT-5.6 Sol
  • GPT-6 Astra
The catalog is curated. A new model shows up in the picker when it is added there. You do not wait on a new MSPilot product for the list to grow. Choose the list in two places. Organization → Power models is the default: All catalog models or a Custom list. If the custom list is empty, power seats see only MSPilot Auto. On the client, Policy → Models is Use organization defaults or Custom list for this client. Power-seat spend is what the MSP is billed. Two limits apply. Each power user is capped at 400perbillingcycle,andalowerper−userlimitcanbesetunderthat.Anoptionalper−clientlimitcoverspowerspendforeverypowerseatintheclient,upto400 per billing cycle, and a lower per-user limit can be set under that. An optional per-client limit covers power spend for every power seat in the client, up to 1,000,000. Leave the client limit blank for no client cap. The per-user cap still applies. Hitting a limit stops the request before it reaches the provider. AI Usage shows usage for the organization. Billing shows what is due this cycle, including power-seat spend. The name on MSPilot Auto is branding, not this model list.
Last modified on September 24, 2026