Skip to main content

​
What are Model Policies?

Model policies control which AI models users in your organization can access. Choose from three configuration modes to define the set of available models. Dashboard: gateway.getunbound.ai/policies/model-policies

​
Configuration Modes

​
All Models

Allow access to every model available on the Unbound gateway. This is the default and most permissive option.

​
Specific Models

Select an explicit list of models that users can access. Only the chosen models will be available — all others are blocked. Use this to restrict usage to approved models only (e.g., only allow gpt-4o and claude-sonnet-4-20250514).

​
All Models with Exclusions

Allow access to all models except a specific set. Use this to block particular models while keeping everything else available. Use this to prevent usage of expensive or unapproved models (e.g., exclude gpt-5.1 to control costs).

​
Setting Up Model Policies

  1. Go to Model Policies in the Unbound dashboard
  2. Choose a configuration mode (All Models, Specific Models, or All Models with Exclusions)
  3. Select or exclude models as needed
  4. Save your configuration

​
Applying to Users

By default, a model policy with no user groups applies to everyone in your organization. To restrict a policy to specific teams, assign it to one or more user groups during creation or editing.

​
Managing via CLI

List and manage model policies from the terminal: 
unbound policy list --type MODEL
unbound policy get <policy-id>

Cost Policies

Set budgets and spending limits

Security Policies

Configure data protection guardrails