Unbabel · Rate Limits

Unbabel Rate Limits

Unbabel does not publish fixed numeric rate limits for the Translation API (tapi/v2) in its public developer documentation. Because translation is an asynchronous, enterprise, quote-based service, throughput is governed in practice by the customer's contracted translation volume/commitment rather than by a documented per-second or per-minute request cap. Submitting a translation returns immediately with a job uid; clients then poll the translation/{uid} endpoint or receive a callback, so polling should be reasonable and backed off rather than tight-looped.

Unbabel Rate Limits is the machine-readable rate-limit profile for Unbabel on the APIs.io network, conforming to the API Commons Rate Limits specification.

It captures 3 rate-limit definitions, measuring requests and words.

The profile also includes 2 backoff/retry policies defined and response codes documented for throttled.

Tagged areas include Translation, LangOps, Rate Limiting, and Quotas.

3 Limits Throttle: 429
TranslationLangOpsRate LimitingQuotas

Limits

Translation API Requests account
requests
not published
No fixed numeric request-rate limit is documented for the tapi/v2 API.
Translation Volume account
words
per contract
Effective throughput is bounded by the customer's committed/quoted monthly translation volume.
Job Status Polling account
requests
not published
Poll translation/{uid} with backoff, or use a callback_url, since jobs complete asynchronously.

Policies

Asynchronous Processing
Translation and mt_translation submissions return a job uid immediately; results are retrieved later by uid or delivered to a callback URL.
Backoff Strategy
Clients should poll job status with exponential backoff and honor Retry-After on any 429 responses.

Sources