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: Vectorize
providerId: vectorize-io
created: '2026-06-20'
modified: '2026-06-20'
reconciled: false
tags:
- AI
- RAG
- Vectorization
- Embeddings
- Retrieval
- Vector Database
- Rate Limiting
- Quotas
- Throttling
description: >-
  The Vectorize API is organization-scoped and Bearer-authenticated. Practical
  usage is governed primarily by subscription-tier quotas - number of RAG
  pipelines, monthly pages processed, and index update frequency - rather than
  granular per-second request ceilings published in the documentation. Specific
  per-endpoint request rate limits for retrieval, extraction, and deep research
  are not reconciled in this artifact and should be confirmed with Vectorize, as
  the API is in Beta and breaking changes may occur.
notes: >-
  Verify request-rate limits, pipeline / page quotas, and any concurrency caps on
  retrieval, Iris extraction, and deep research in the Vectorize platform console
  and developer docs on reconciliation; quotas scale with the Free / Starter /
  Professional / Enterprise tier.
sources:
- https://docs.vectorize.io/reference/api/api-getting-started/
- https://docs.vectorize.io
- https://vectorize.io/
responseCodes:
  throttled: 429
limits:
- name: RAG Pipelines
  scope: organization
  metric: pipelines
  limit: see plan (1 Free / 3 Starter / 10 Professional / custom Enterprise)
  notes: Maximum concurrent RAG pipelines per organization, set by subscription tier.
- name: Pages Processed
  scope: organization
  metric: pages
  limit: see plan (1,500/month Free; higher on paid tiers)
  notes: Monthly document page-processing allowance for ingestion / extraction.
- name: Index Update Frequency
  scope: pipeline
  metric: schedule
  limit: see plan (hourly on Professional)
  notes: How often a pipeline refreshes its vector index, gated by tier.
- name: Retrieval Requests
  scope: organization
  metric: requests
  limit: see provider documentation
  notes: Per-pipeline retrieval query volume; specific rate not documented.
- name: Extraction Jobs
  scope: organization
  metric: jobs
  limit: see provider documentation
  notes: Iris extraction jobs run asynchronously; poll the result endpoint until ready.
- name: Deep Research Runs
  scope: organization
  metric: jobs
  limit: see provider documentation
  notes: Deep research runs are asynchronous; poll for the report by researchId.
policies:
- name: Tiered Quotas
  description: Pipeline counts, page allowances, and update frequency raise as accounts move from Free to Starter to Professional to Enterprise.
- name: Asynchronous Polling
  description: Extraction and deep research are async; clients poll the result endpoint until ready is true rather than holding a long connection.
- name: Backoff Strategy
  description: Clients should implement exponential backoff with jitter on 429 responses and honor Retry-After when present.
- name: Beta Stability
  description: The API is in Beta; breaking changes may occur, so clients should pin client-library versions and monitor changelogs.
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com