Amazon Comprehend vs spaCy
No leader: the top candidate Amazon Comprehend has only 0.33 confidence (low), below the 0.35 needed to declare a winner. The attribute-by-attribute breakdown below, with a source and date on every value, is the honest way to compare them.
Capabilities
Feature-by-feature on the axes that matter for nlp apis. “-” means undocumented, not absent.
What each one is
The product in its own terms, so the numbers below have context.
Amazon Comprehend
A machine learning service that processes text to identify entities, sentiment, topics, and key phrases. It supports both standard pre-built models and custom models trained on your own data, with APIs for document processing in multiple formats.
spaCy
An open-source natural language processing library designed for building real-world applications. It excels at large-scale information extraction and has become an industry standard since 2015 with a robust ecosystem of plugins.
Pricing
List pricing as published by each vendor, with the date we read it. Always verify at the source before you buy.
Amazon Comprehend
From $0.0001 per unit (100 characters) for standard APIs. Free tier: 50K units per month for 12 months.
- Standard APIs (Pay-per-use)$0.0001 per unit (100 characters)
- Entity recognition
- Sentiment analysis
- Syntax analysis
- Key phrase extraction
- Language detection
- +5 more
- Custom Classification & Entities$3/hour training + $0.50/month management + $0.0003/unit inference
- Custom model training
- Custom entity extraction
- Custom text classification
- Model management
- Managed Endpoints (Synchronous)$0.0005 per second per Inference Unit
- Real-time synchronous inference
- Scalable throughput with multiple IUs
Platform & deployment
Where each product runs and how it can be hosted. A dash means undocumented, not unsupported.
Integrations
What each product connects to. Counts come from the vendor's own integration directory where one exists.
Amazon Comprehend
Not documented yet.
spaCy
- PyTorch
Comparison generated from independently-sourced facts. Every value links to its source and retrieval date. See the method.