Amazon Translate vs Lara Translate API

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.

Lara Translate API

Glossaries, translation memories, styles and instructions on every translate call. Texts may be used for model improvement unless a request sets noTrace.

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

Lara Translate API

1. Set noTrace on every request that carries private text, since storage for model improvement is the default
2. Leave reasoning off unless the text needs it. It moves the call to Lara Think at $2,499 per million characters
3. Resolve glossary and memory names to gls_* and mem_* IDs with the list tools before translating
4. Send one target language per translate call
5. Merge short documents, since each file bills at least 20,000 characters