Skip to content
SocialAtoZ

Amazon Comprehend

Verified

AWS managed NLP service for sentiment, entities, and custom text classification

Text Analysis Software
–

Not yet rated. Be the first to review Amazon Comprehend.

Visit website Request pricing Write a review
Amazon Comprehend screenshot See all screenshots
  • Deployment Cloud Based
  • Starting price Free
  • Free trial Available
  • Best for Startups, SMEs, Enterprises

What is Amazon Comprehend?

Amazon Comprehend is an AWS managed natural language processing service that uses machine learning to find insights in text without requiring users to build models from scratch. It offers sentiment analysis, entity recognition, key phrase extraction, language detection, syntax analysis, PII detection and redaction, and toxicity detection through simple API calls. Custom classification and custom entity recognition let teams train their own models on labeled data using AutoML.

Pricing is pay as you go, billed per 100-character unit, with a free tier covering 50,000 units per API each month for the first 12 months. Custom model training and hosted inference endpoints are billed separately by the hour or second. Comprehend targets developers and enterprises already using AWS who want to add text analytics to applications, pipelines, or support workflows.

Key Features of Amazon Comprehend

Amazon Comprehend lists 9 documented features, including Sentiment analysis (positive, neutral, negative, mixed), Entity recognition and Key phrase extraction. The list below covers what the product does rather than how it is marketed.

  • Sentiment analysis (positive, neutral, negative, mixed)
  • Entity recognition
  • Key phrase extraction
  • Language detection
  • Syntax analysis
  • PII detection and redaction
  • Toxicity and prompt safety detection
  • Custom classification and custom entity recognition via AutoML
  • Topic modeling for document collections

Amazon Comprehend Pricing

Amazon Comprehend lists 4 plans. One is free; the rest are quoted on request.

Free Tier (12 months)

Free

50,000 units per API per month

Request pricing

Available to new and existing AWS customers for the first 12 months from first request

Standard pay-as-you-go

$0.0001/unit

300-character minimum per request

Request pricing

Covers standard NLP APIs like sentiment, entities, and key phrases

Custom model training

$3/hour

Billed per second

Request pricing

Plus $0.50/month model management and $0.0005 per inference-unit-second for hosted endpoints (60-second minimum)

Topic Modeling

$1.00 flat

(first 100MB)

Per job

Request pricing

$0.004 per MB for document volume above 100MB

Amazon Comprehend Specifications

Amazon Comprehend is available on web app. It offers an API, has a free trial and has a free plan.

Deployment
  • Cloud Based
Billing cycle
Monthly
Desktop
  • Web App
Languages
  • English
Built for
  • Startups
  • SMEs
  • Enterprises
Support
  • Email
  • Tickets
  • Phone
  • Live Support
Integrations
AWS services including S3, Lambda, SageMaker, Kinesis, and Lex; AWS SDKs for Python (boto3), Java, Node.js, .NET, and Go
Public API
Yes
Free trial
Yes
Free plan
Yes
Runs in browser
Yes
Customisable
Yes
Website
aws.amazon.com

Amazon Comprehend Videos

Amazon Comprehend Screenshots

Amazon Comprehend screenshot 1

Used Amazon Comprehend? Write the first review.

Amazon Comprehend FAQs

Yes, new and existing AWS customers get 50,000 units per API per month free for 12 months for most standard APIs.

Yes, Comprehend includes Detect PII and Contains PII APIs that identify and can redact personally identifiable information.

Yes, custom classification and custom entity recognition use AutoML to train models on your own examples, with training billed at $3 per hour.

Comprehend supports multiple languages including English, Spanish, French, German, Italian, Portuguese, Arabic, Hindi, Japanese, Korean, and Chinese, though not every feature covers every language.

Standard APIs are billed per 100-character unit with a 300-character minimum per request, plus separate charges for custom model training, management, and inference endpoints.