Menu
Amazon Comprehend
Developer Guide

Guidelines and Limits

Guidelines

Amazon Comprehend is available in the following regions:

Region Name Region Endpoint Protocol
US East (Ohio) us-east-2 https://comprehend.us-east-2.amazonaws.com HTTPS
US East (N. Virginia) us-east-1 https://comprehend.us-east-1.amazonaws.com HTTPS
US West (Oregon) us-west-2 https://comprehend.us-west-2.amazonaws.com HTTPS
EU (Ireland) eu-west-1 https://comprehend.eu-west-1.amazonaws.com HTTPS

Limits

Amazon Comprehend has the following limits:

  • Documents submitted for topic detection using the StartTopicsDetectionJob operation must be UTF-8 encoded text files.

  • The maximum document size is 5,000 bytes of UTF-8 encoded characters.

  • The maximum number of documents for the BatchDetectDominantLanguage, BatchDetectEntities, BatchDetectKeyPhrases, and BatchDetectSentiment operations is 25 documents per request.

  • The BatchDetectDominantLanguage and DetectDominantLanguage operations have the following limitations:

    • They don't support phonetic language detection. For example, they will not detect "arigato" as Japanese, nor "nihao" as Chinese.

    • They may have trouble distinguishing close language pairs, such as Indonesian and Malay; or Bosnian, Croatian, and Serbian.

    • For best results, the input text should be at least 20 characters long.

  • Topic detection jobs created with the StartTopicsDetectionJob operation have the following limits:

    • For best results, you should include at least 1,000 input documents.

    • The maximum number of topics to return can't exceed 100.

    • The total size of all files in the request can't exceed 1 Gb.

    • The total size of a single file in the request can't exceed 2 Mb.

    • When you are using one document per file, you can send a maximum of 50,000 files.

    • When you are using one document per line, you can send a maximum of 1,000,000 lines.

    To request a service limit increase, see https://aws.amazon.com/support..

On this page: