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: Sonix
providerId: sonix
created: '2026-07-11'
modified: '2026-07-11'
reconciled: false
tags:
- Audio Transcription
- Transcription
- Speech-to-Text
- Rate Limiting
- Quotas
description: >-
  Sonix does not publish fixed numeric request-rate limits for its REST API.
  Practical throughput is governed by two things: your plan's transcription /
  translation and AI-workspace hour allowances (per-hour on pay-as-you-go, or
  the bundled monthly hours on Core/Advanced/Pro), and per-request constraints
  such as the 100MB multipart upload size (larger files must be supplied via a
  file_url). Media listing is paginated at 100 items per page. Transcription is
  asynchronous, so clients should poll media/transcript status or use a
  completion webhook rather than issuing tight request loops.
notes: >-
  Numeric per-account or per-endpoint request limits are not documented as of
  the review date. The dominant "limit" is your plan's hour allowance and
  storage quota. Verify any published limits in the Sonix API docs during
  reconciliation.
sources:
- https://sonix.ai/docs/api
- https://sonix.ai/api
- https://sonix.ai/pricing
responseCodes:
  throttled: 429
limits:
- name: API Request Rate
  scope: account
  metric: requests
  limit: not published
  notes: No fixed numeric per-minute request-rate limit is documented for the REST API.
- name: Transcription / Translation Hours
  scope: account
  metric: hours
  limit: per plan
  notes: Per-hour on pay-as-you-go ($10/hr); bundled monthly hour allowances on Core/Advanced/Pro.
- name: AI Workspace Hours
  scope: account
  metric: hours
  limit: per plan
  notes: AI analysis (summaries, chapters, sentiment, entities) draws down a separate monthly AI-workspace hour allowance.
- name: Multipart Upload Size
  scope: request
  metric: bytes
  limit: 104857600
  notes: Multipart file uploads are limited to 100MB; larger or remote files must be passed via file_url.
- name: Media List Pagination
  scope: request
  metric: items
  limit: 100
  notes: The media list endpoint returns up to 100 items per page and includes total_pages.
- name: Storage
  scope: account
  metric: bytes
  limit: per plan
  notes: Storage quota ranges from 5 GB (pay-as-you-go) to 100 GB (Pro) by plan.
policies:
- name: Asynchronous Processing
  description: Transcription, translation, exports, and AI analysis are asynchronous jobs; retrieve results by polling status or via a completion webhook (callback_url).
- name: Backoff Strategy
  description: Clients should implement exponential backoff with jitter and honor Retry-After on 429 responses rather than polling aggressively.
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com