Published Oct 15, 2026
| Starting price (indicative) | Published examples historically referenced per-asset and per-minute video pricing; costs vary by feature group and region. | Published examples show per-unit pricing for many features with a free tier for limited units; per-feature billing applies. | Published pricing exists for Azure Cognitive Services; pricing varies by region and selected feature set. |
| Fit | Integrates with AWS-native storage, identity, and orchestration services. | Preferred by organizations using GCP; comprehensive per-feature pricing for many capabilities. | Preferred by organizations standardized on Microsoft platforms and governance tooling. |
| Key strengths | Wide set of API capabilities for images and video with integration to AWS services. | Explicit per-feature pricing and a broad set of vision features. | Enterprise alignment for Microsoft environments and procurement processes. |
| Key weaknesses | Unit economics at scale for face matching and continuous video can be unfavorable; reputational and regulatory concerns for facial recognition; feature deprecations have occurred. | Costs can accumulate when multiple features are applied per image because many features are billed separately. | Pricing and navigation complexity for some cognitive services; potential for unexpected costs depending on feature usage patterns. |
—
Where Rekognition aligns with buyer needs:
Where competitors often align better:
Market-share and adoption metrics from third-party sites vary in methodology; such figures should be treated cautiously for procurement decisions.