Provider Allowlist
The provider allowlist lets team owners restrict which AI providers can serve requests through AI Gateway. A request only returns a 403 when no allowed provider can serve it. The feature is opt-in and available on Pro and Enterprise plans.
A common reason to enable this is compliance: your team has reviewed a specific set of providers, and you want to guarantee that AI Gateway never routes to one you haven't approved.
The provider allowlist is a team-wide setting that applies to every request. If you only need to restrict providers on individual requests, use the only parameter in providerOptions instead at no additional cost.
| Option | Cost | Availability |
|---|---|---|
Per-request only filter | No additional cost | All plans |
| Team-wide provider allowlist | $0.10 per 1,000 successful requests | Pro and Enterprise |
The team-wide surcharge applies only to successful responses. Requests blocked by the allowlist (403) and other failures are not charged.
Only team owners can change this setting.
- Open the AI Gateway Settings tab.
- Toggle on Restrict allowed providers.
When you toggle on, every provider currently available on AI Gateway is added to the allowlist by default. This guarantees that turning the feature on never breaks existing traffic. You then remove providers you want to block as an explicit action.
With the allowlist enabled, switch off any providers your team should not use. The list is searchable. The footer shows how many providers are currently enabled.
Disabling a provider removes it from the routing candidates for any request. If the model has other providers that are still allowed, AI Gateway falls back to one of them. If every candidate provider for the request is disabled, the request returns:
{
"error": "Your team has restricted access to this provider. Contact the owner of the account for more details.",
"type": "no_providers_available",
"statusCode": 403
}The allowlist filters by provider only. A model like openai/gpt-5.4-mini is served by both openai and azure; disabling openai alone still routes that model through azure. To block a model entirely, disable every provider that serves it.
When AI Gateway adds a new provider after you enable the allowlist, the new provider is disabled by default and shows a New badge in the settings list. An info note at the top of the list tells you how many new providers are waiting for a decision.
To clear the badges, decide which new providers to enable and click Save. A save with no changes also counts as acknowledgment and clears the badges. Disabled new providers stay disabled until you enable them.
If you also use the request-level only filter in providerOptions, both filters apply. A request must satisfy both:
- The provider must be in the request's
onlylist, and - The provider must be in your team's allowlist.
If either filter rejects every candidate, the request returns a 403.
The allowlist applies to BYOK requests too. Disabling a provider means your team does not want any traffic to that provider through AI Gateway, even with your own credentials.
| Requirement | Detail |
|---|---|
| Plan | Pro or Enterprise (not Pro trial) |
| Role | Owner |
Team members on non-owner roles can view the current configuration but cannot modify it.
Was this helpful?