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: Voxel51
providerId: voxel51
created: '2026-06-21'
modified: '2026-06-21'
reconciled: false
tags:
- AI
- Computer Vision
- Datasets
- Machine Learning
- Python SDK
- Rate Limiting
- Quotas
description: >-
  FiftyOne's primary interface is a local, open-source Python SDK, so there is
  no public per-request rate limit on the open-source surface - throughput is
  bounded by the user's own hardware and the backing MongoDB/storage. FiftyOne
  Enterprise capacity is governed by contracted entitlements (user seats, VPUs,
  and monthly compute hours) rather than published RPM/TPM API throttles, and
  the authenticated API connection is sized by the customer's deployment. No
  public numeric rate limits are documented; values are not reconciled here.
notes: >-
  Open-source FiftyOne has no remote API and therefore no provider rate limit.
  Enterprise limits are effectively the contracted seat / VPU / compute-hour
  entitlements plus any deployment-level capacity; confirm with Voxel51 during
  reconciliation.
sources:
- https://docs.voxel51.com/
- https://docs.voxel51.com/enterprise/api_connection.html
- https://voxel51.com/pricing
responseCodes:
  throttled: 429
limits:
- name: Open Source SDK
  scope: local
  metric: requests
  limit: not applicable
  notes: Local Python library; throughput bound by user hardware and backing database, not a provider quota.
- name: Enterprise User Seats
  scope: account
  metric: seats
  limit: see contract
  notes: Entitled seats per tier (e.g., 8 Team / 25 Growth / unlimited Custom).
- name: Enterprise VPUs
  scope: account
  metric: vpus
  limit: see contract
  notes: Visual processing units provisioned per tier.
- name: Enterprise Compute Hours
  scope: account
  metric: compute_hours
  limit: see contract
  notes: Monthly compute-hour entitlement per tier; not a per-minute API throttle.
policies:
- name: Entitlement-Based Capacity
  description: Enterprise usage is governed by contracted seats, VPUs, and compute hours rather than published API rate limits.
- name: Backoff Strategy
  description: SDK clients hitting the Enterprise API connection should implement retries with exponential backoff and honor Retry-After if returned.
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com