Martian · Rate Limits

Martian Ai Rate Limits

The Martian Gateway is a routing layer in front of many upstream provider models, so effective throughput is governed both by Martian account limits and by the rate limits of whichever provider model the router selects. Martian documents a free request allotment and returns standard HTTP 429 responses when limits are exceeded, but specific per-account RPM/TPM values are not publicly enumerated and are not reconciled here.

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

It captures 4 rate-limit definitions, measuring requests and tokens.

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

Tagged areas include AI, LLM, Model Router, Gateway, and Cost Optimization.

4 Limits Throttle: 429
AILLMModel RouterGatewayCost OptimizationRate LimitingQuotasThrottling

Limits

Requests Per Minute (RPM) account
requests
see provider documentation
Per-account request rate; not publicly enumerated.
Tokens Per Minute (TPM) account
tokens
see provider documentation
Per-account token throughput; not publicly enumerated.
Free Request Allotment account
requests
2500
Documented free developer allotment before metered paid usage applies.
Downstream Provider Limits model
requests
varies by routed provider
The selected upstream model's own provider rate limits may constrain a routed request.

Policies

Tiered Limits
Limits raise from the free developer tier to paid and Enterprise agreements.
Backoff Strategy
Clients should implement exponential backoff with jitter and honor Retry-After on 429 responses.
Automatic Failover
Martian can fail over across providers, which can mitigate single-provider throttling for a routed request.

Sources