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: DataCrunch
providerId: datacrunch
created: '2026-06-21'
modified: '2026-06-21'
reconciled: false
tags:
- GPU Cloud
- Infrastructure
- Compute
- Inference
- Serverless
- Europe
- Rate Limiting
- Quotas
- Throttling
description: >-
  The DataCrunch Public API at https://api.datacrunch.io/v1 is gated by OAuth2
  Bearer access tokens scoped to cloud-api-v1. Practical limits are driven by
  resource quotas (available GPU capacity per location, account spend/credit
  balance, and per-project instance/volume allowances) rather than published
  fixed request-per-minute numbers. Serverless container deployments enforce
  autoscaling bounds via min/max replica counts and queue limits per
  deployment. Specific numeric API request rate limits are not publicly
  documented and are not reconciled in this artifact.
notes: >-
  Verify any per-endpoint request throttles, token expiry/refresh behavior,
  and per-account GPU/instance quotas in the live API reference and account
  console on reconciliation. Capacity-driven 4xx responses (no availability)
  are distinct from request throttling.
sources:
- https://api.datacrunch.io/v1/docs
- https://docs.datacrunch.io/
- https://docs.datacrunch.io/containers/scaling
responseCodes:
  throttled: 429
limits:
- name: API Requests
  scope: account
  metric: requests
  limit: see provider documentation
  notes: No fixed public RPM/RPD published; requests authorized via OAuth2 Bearer token.
- name: GPU Capacity / Availability
  scope: account
  metric: instances
  limit: subject to real-time availability per location
  notes: Deploys can be rejected when the requested instance type has no availability.
- name: Account Quota / Balance
  scope: account
  metric: credit
  limit: gated by prepaid balance and project quotas
  notes: New deployments require sufficient account balance/credit.
- name: Deployment Replicas
  scope: deployment
  metric: replicas
  limit: min/max replica count per serverless deployment
  notes: Autoscaling bounded by configured min/max replicas and queue length per replica.
- name: Deployment Queue
  scope: deployment
  metric: messages
  limit: per-deployment request queue with message TTL
  notes: Queue can be purged via the purge-queue endpoint; messages expire per configured TTL.
policies:
- name: OAuth2 Token Scope
  description: Tokens issued via client-credentials are scoped to cloud-api-v1 and restricted to documented Public API endpoints; refresh before expiry.
- name: Capacity-Aware Retry
  description: Treat no-availability responses as capacity signals; retry against alternate instance types or locations rather than tight-looping.
- name: Backoff Strategy
  description: Clients should implement exponential backoff with jitter and honor Retry-After on 429 responses.
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com