Amazon Translate vs Crowdin

Amazon Translate

IAM policies scope access per action and resource, with a TranslateReadOnly managed policy, and CloudTrail logs each call. AWS may store and use text to improve Amazon AI services unless the organisation sets an opt-out policy.

Crowdin

The REST API has a public OpenAPI document with 374 operations, and tokens can be limited by scope, access level and project. Translation runs inside Crowdin projects and connected engines, so it is not a stateless per-character translation API. No Retry-After guidance, idempotency keys or SLA were found in the reviewed documentation.

Amazon Translate

1. Chunk real-time text under 10,000 bytes, not characters, so multibyte scripts fit less per call
2. Pass SourceLanguageCode auto to detect the source language, and handle DetectedLanguageLowConfidenceException
3. Read AppliedSettings in the response, since unsupported formality or profanity settings are dropped without an error
4. Retry TooManyRequestsException with backoff, which the AWS SDKs do by default
5. Set the AI services opt-out policy before sending private text

Crowdin

1. Send Authorization: Bearer <token> to https://api.crowdin.com/api/v2. Crowdin Enterprise uses https://<organization_domain>.api.crowdin.com/api/v2
2. Use the MCP V2 endpoint https://mcp.crowdin.com/v2/mcp and set crowdin-tool-sets, because the server loads all 33 tool sets by default
3. Set crowdin-read-only: true on the MCP connection for exploration. It exposes only list and get tools
4. Keep to 20 concurrent requests an account across all tokens, and add your own backoff on 429
5. Upload a file to /storages first, then pass the storageId to the add-file call. Builds, exports and reports are asynchronous and need a status poll
6. /mts/{mtId}/translations takes at most 100 strings a call and needs the mt scope and a configured engine