name: Dataiku API Rate Limits
specification: API Commons Rate Limits
specificationVersion: '0.1'
schema: https://raw.githubusercontent.com/api-evangelist/interface-research/main/schema/api-commons.yml#/$defs/RateLimits
provider: Dataiku
providerId: dataiku
created: '2026-05-04'
# Provenance stamped 2026-08-11: this artifact was written by the API Evangelist
# bulk sweep dated 2026-05-04, not harvested from the provider. See roadmap#35.
method: generated
modified: '2026-05-05'
reconciled: false
tags:
  - Analytics
  - Artificial Intelligence
  - Data Platform
  - Machine Learning
  - Rate Limiting
description: Dataiku DSS is typically self-hosted (or hosted in a customer's cloud tenant), so
  there are no global vendor-imposed rate limits on the Public API or Internal API. Throughput is
  bounded by the DSS backend node CPU/memory, the API Node deployment for real-time scoring, and
  per-API-service quotas defined by the operator. The Cloud Trial exposes a fixed compute envelope
  (4 CPUs, 32 GB) which effectively caps trial throughput.
sources:
  - https://www.dataiku.com/product/get-started/
  - https://doc.dataiku.com/
notes: No public numeric rate limits. Operators define rate-limit middleware per API Node service
  if needed.
responseCodes:
  throttled: 429
  serviceUnavailable: 503
limits:
  - name: Cloud Trial compute envelope
    scope: trial-tenant
    metric: varies
    limit: 4 CPUs and 32 GB elastic compute
    notes: Trial throughput is capped by allocated CPU/memory rather than a request-per-second
      ceiling.
  - name: DSS backend capacity
    scope: deployment
    metric: requests_per_second
    limit: operator-defined — bounded by backend node sizing
    notes: Self-hosted; the Public and Internal APIs throughput depends on the DSS node sizing.
  - name: API Node service quotas
    scope: per-service
    metric: requests_per_second
    limit: operator-defined per API Service
    notes: Real-time scoring services on the API Node can be configured with per-service rate
      limits and concurrency caps via the Administration API.
policies:
  - name: API Node Sizing
    description: For real-time scoring, scale API Node replicas horizontally and use a load
      balancer; per-service concurrency is controlled in the API Service deployment configuration.
  - name: Backoff Strategy
    description: Clients should retry on 429/503 with exponential backoff and jitter.
  - name: Public API vs Internal API
    description: The Public API is the supported integration surface for Projects, Datasets, and
      Workflows. The Internal API is undocumented and may change without notice; use sparingly.
  - name: Bulk Operations
    description: Prefer bulk endpoints (multi-row scoring, bulk dataset write) over per-row calls
      to reduce request volume.
maintainers:
  - FN: Kin Lane
    email: kin@apievangelist.com
    url: https://www.dataiku.com