Maven AGI · Rate Limits

Maven Agi Rate Limits

The Maven AGI Platform API applies per-app and per-agent throttling, but specific numeric limits are not published in the public documentation and are governed by the customer's enterprise agreement. The official SDKs (Python, Node, Java, Go, .NET) ship with automatic retries and exponential backoff, retrying requests deemed retryable up to a default of two attempts, which signals that clients are expected to handle throttling gracefully.

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

It captures 3 rate-limit definitions, measuring requests and connections.

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

Tagged areas include AI, Agents, Customer Support, Rate Limiting, and Quotas.

3 Limits Throttle: 429
AIAgentsCustomer SupportRate LimitingQuotasThrottling

Limits

Per-App Requests app
requests
see provider documentation
Throttling is scoped to the App ID / App Secret credential; values not published.
Per-Agent Requests agent
requests
see provider documentation
Requests are scoped to X-Agent-Id; limits governed by the customer agreement.
Streaming (ask_stream) agent
connections
see provider documentation
Server-Sent Events streaming for ask_stream; concurrency limits not published.

Policies

Automatic Retries
Official SDKs retry retryable requests with exponential backoff (default 2 retries).
Backoff Strategy
Clients should implement exponential backoff with jitter and honor Retry-After on 429 responses.
Enterprise-Governed Limits
Concrete quotas are set per enterprise agreement rather than published as fixed public tiers.

Sources