Langfuse · Rate Limits

Langfuse Rate Limits

Langfuse cloud throttling is primarily quota-based on monthly units (traces, observations, scores). Enterprise accounts can negotiate custom rate limits. Self-hosted installations are limited only by the operator's own infrastructure.

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

It captures 4 rate-limit definitions, measuring unit, varies, and none.

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

Tagged areas include AI, LLM, Observability, Open Source, and Evaluations.

4 Limits Throttle: 429
AILLMObservabilityOpen SourceEvaluationsRate LimitingQuotasThrottling

Limits

Hobby Cloud Monthly Units account
unit · monthly
50000
Hard cap for the free Hobby tier.
Core/Pro/Enterprise Monthly Units (Included) account
unit · monthly
100000
Included monthly allowance; overage billed at $8/100k with volume discounts.
Enterprise Custom Rate Limits account
varies
contract
Enterprise customers may negotiate custom per-second or per-minute API rate ceilings.
Self-Hosted instance
none
-1
No platform-imposed limits when self-hosted; bounded by operator infrastructure.

Policies

Backoff Strategy
Honor 429 responses with exponential backoff and jitter; SDKs handle retry automatically.
Trace Batching
SDKs batch trace ingestion calls before flushing, reducing per-request overhead.
OpenTelemetry Path
Use the OTel ingestion endpoint where possible; it is the recommended, higher-throughput path.

Sources