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: Phraseanet
providerId: phraseanet
created: '2026-07-05'
modified: '2026-07-05'
reconciled: false
tags:
- Digital Asset Management
- DAM
- Open Source
- Rate Limiting
- Quotas
description: >-
  Phraseanet does not impose a fixed API rate limit. The public v1 API
  documentation states explicitly that there is no rate limit in v1 - you may
  make as many requests as you wish. Because Phraseanet is self-hosted, real
  throughput is bounded by the resources of the instance you run (PHP-FPM
  workers, database, and the Elasticsearch cluster that powers search) rather
  than by a vendor-imposed per-minute cap. Operators can apply their own limits
  at the reverse proxy or web server in front of Phraseanet.
notes: >-
  No numeric per-account or per-endpoint limit is published by Phraseanet.
  Search-heavy workloads are gated by Elasticsearch capacity. Tune capacity and
  any protective throttling at your own infrastructure layer.
sources:
- https://docs.phraseanet.com/4.1/en/Devel/API/V1/Overview.html
- https://docs.phraseanet.com/4.1/en/Devel/API/V1.html
- https://github.com/alchemy-fr/Phraseanet
responseCodes:
  throttled: 429
limits:
- name: API Requests (v1)
  scope: instance
  metric: requests
  limit: no rate limit (per v1 documentation)
  notes: v1 documentation states there is no rate limit; requests are unbounded by Phraseanet.
- name: Self-Hosted Throughput
  scope: deployment
  metric: requests
  limit: hardware-bound
  notes: Constrained by your own PHP, database, and Elasticsearch capacity, not by Phraseanet.
- name: Search Throughput
  scope: instance
  metric: queries
  limit: elasticsearch-bound
  notes: Search is served by Elasticsearch; throughput depends on cluster sizing.
- name: Operator-Imposed Limits
  scope: instance
  metric: requests
  limit: configurable
  notes: Operators may add rate limiting at a reverse proxy or web server in front of Phraseanet.
policies:
- name: No Vendor Rate Limit
  description: Phraseanet does not enforce an API rate limit in v1; throughput is a function of your own deployment.
- name: Backoff Strategy
  description: Clients should still implement exponential backoff with jitter and honor Retry-After should an operator-imposed proxy return 429.
- name: Capacity Planning
  description: Scale PHP workers, the database, and the Elasticsearch cluster to match expected API and search load.
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com