Fern · Rate Limits

Fern Rate Limits

Fern's customer-facing limits are expressed as plan-level metering rather than per-second HTTP rate limits. Public Fern surfaces — the docs site, dashboard, and the fern-mcp-server hosted endpoint — do not document per-request rate limits on their public pages, so this file captures the consumption metering that does exist (seats and AI credits) plus generic safe-harbor guidance for programmatic consumers.

Fern Rate Limits is the machine-readable rate-limit profile for Fern on the APIs.io network, conforming to the API Commons Rate Limits specification.

It captures 5 rate-limit definitions, across the hobby, team, and enterprise tiers, measuring seats and ai_credits_per_month.

The profile also includes 3 backoff/retry policies defined and response codes documented for throttled, quotaExceeded, and serviceUnavailable.

Tagged areas include AI, API Definitions, Code Generation, Developer Experience, and Documentation.

5 Limits Throttle: 429 Quota: 429
AIAPI DefinitionsCode GenerationDeveloper ExperienceDocumentationOpenAPIPlatformSDKsRate LimitingQuotas

Limits

Hobby Seats workspace
seats · month
2
Hobby AI Credits workspace
ai_credits_per_month · month
250
Team Seats workspace
seats · month
5
Team AI Credits workspace
ai_credits_per_month · month
1000
Enterprise Negotiated contract
seats · month
-1

Policies

Backoff Strategy
MCP and dashboard clients should retry with exponential backoff and respect any Retry-After header on 429 or 503 responses.
Fair Use
Even on enterprise tiers, sustained or abusive usage that impacts shared infrastructure (e.g. the hosted MCP endpoint) may be throttled.
AI Credit Reset
AI credits reset at the start of each billing cycle.