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: SYSTRAN
providerId: systran
created: '2026-07-03'
modified: '2026-07-03'
reconciled: false
tags:
- Machine Translation
- Translation
- NLP
- Rate Limiting
- Quotas
description: >-
  SYSTRAN does not publish a fixed public per-minute request-rate limit for the
  Translate API. Practical throughput is governed by the character allowance of
  your subscription (metered per character, starting with the 500,000-character
  free trial) rather than by a documented requests-per-second cap. Large or
  long-running documents are handled through the asynchronous file translation
  flow (async=true) and batches, which decouple job submission from a per-request
  size limit. Per-request text size, concurrency, and account-level throughput
  ceilings are enforced by SYSTRAN but are not published as fixed numeric values.
notes: >-
  No numeric per-account or per-endpoint request-rate limits are documented as of
  the review date. Use asynchronous file translation and batches for large volumes,
  implement exponential backoff on 429 responses, and confirm any per-plan
  character allowances and concurrency limits with SYSTRAN during reconciliation.
sources:
- https://docs.systran.net/translateAPI/translation/
- https://docs.systran.net/translateAPI/en/
- https://www.systransoft.com/translation-products/translate-api/
responseCodes:
  throttled: 429
limits:
- name: Translate API Requests
  scope: account
  metric: requests
  limit: not published
  notes: No fixed numeric request-rate limit is documented for the Translate API.
- name: Character Allowance
  scope: account
  metric: characters
  limit: per subscription (500,000 on the free trial)
  notes: Usage is metered per character; throughput is bounded by your plan's character allowance.
- name: Synchronous Text Size
  scope: request
  metric: characters
  limit: bounded per request
  notes: Very large content should use asynchronous file translation instead of a single synchronous text call.
- name: Asynchronous File Jobs
  scope: account
  metric: jobs
  limit: managed via batches
  notes: Large or long-running documents are submitted with async=true and polled by requestId; batches group multiple requests.
policies:
- name: Asynchronous Processing
  description: Submit large documents with async=true and poll /translation/file/status by requestId; group multiple jobs into a batch.
- name: Backoff Strategy
  description: Clients should implement exponential backoff with jitter and honor Retry-After on 429 responses.
- name: Character Metering
  description: Billing and effective throughput are driven by characters translated rather than a per-second request cap.
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com