Superlinked · Rate Limits

Superlinked Rate Limits

The Superlinked Server is self-hosted open-source software and does not enforce vendor-imposed rate limits; throughput is bounded by the compute you provision and the limits of the external vector database and embedding model inference behind it. Any rate limiting in front of the generated REST API is the operator's responsibility (gateway, reverse proxy, or custom middleware). Superlinked Cloud may apply plan-specific limits, but none are publicly documented. No specific limits are reconciled in this artifact.

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

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

The profile also includes 2 backoff/retry policies defined and response codes documented for throttled.

Tagged areas include Vectors, Embeddings, Vector Search, Retrieval, and Recommendations.

4 Limits Throttle: 429
VectorsEmbeddingsVector SearchRetrievalRecommendationsRAGRate LimitingQuotasThrottling

Limits

Server Throughput deployment
requests
bounded by provisioned compute
Self-hosted; no vendor-imposed RPM/TPM. Scale horizontally to raise throughput.
Vector Database Limits deployment
operations
see backing vector DB
Effective limits inherited from Redis / MongoDB / Qdrant / TopK and your plan there.
Embedding Inference deployment
embeddings
bounded by inference capacity
Throughput of embedding model inference (local or provider) constrains ingestion and query rates.
Superlinked Cloud account
requests
see provider documentation
Managed cloud is early access / sales-led; plan limits are not publicly documented.

Policies

Operator-Enforced Limiting
Place rate limiting at a gateway or proxy in front of the self-hosted server; the server itself does not throttle.
Backoff Strategy
Clients should implement exponential backoff with jitter and honor Retry-After if a fronting proxy returns 429.

Sources