# Amazon Bedrock — AI Compare Ninja

> Managed AWS platform for AI models, knowledge bases and agents with usage-based pricing.

## Overview
- **Category:** Development & Code (Agent & RAG frameworks)
- **Status:** Active · checked - Active.
- **Paid from:** Not verified for this market / Usage (On-demand / Standard)
- **Free access:** One-off free allowance (Only eligible new AWS customers qualify: USD 100 in starting credits, plus up to USD 100 more through activities. These are shared AWS credits, not a permanent free Bedrock plan. The free account ends after 6 months or when the credit runs out.)
- **Website:** https://aws.amazon.com/bedrock/
- **Canonical URL:** https://www.aicompare.ninja/en/tool/amazon-bedrock

## What it does
- Models from Amazon, Anthropic, Meta, Mistral and other providers
- Standard, Priority, Flex, Batch and reserved capacity depending on model
- In-region, geographic and global inference profiles differ in processing location

## In brief
Managed AWS platform for AI models, knowledge bases and agents with usage-based pricing. Prices depend on model, modality, AWS Region and service tier. Text models may charge for input/output tokens, image services per generation and video models by duration. Provisioned capacity, storage, Guardrails, Knowledge Bases and other services may add charges. AWS starting credits do not imply a fixed Bedrock token allowance.


## Limits
Region boundaries depend on the inference profile: Geographic Cross-Region routes within the selected geography such as the EU or US; Global Cross-Region can use supported commercial AWS Regions worldwide. A European entry point alone therefore does not guarantee processing in that exact Region. AWS says model providers cannot access Bedrock model deployment accounts, prompts or completions; retention, abuse detection and your connected services still need separate consideration.

## Pricing tiers
| Tier | Price | Period | What you get |
| --- | --- | --- | --- |
| AWS Free Tier credits | free | One-time starting credits | Eligible new AWS customers only: USD 100 starting credits, up to USD 100 more through activities; Shared AWS credits, not a permanent free Bedrock plan; free account plan ends after 6 months or credit exhaustion |
| On-demand / Standard | Not verified for this market | Usage | No single platform price; charges depend on model, modality and AWS Region; Units vary by feature: tokens, images, audio/video duration or other measures |
| Priority | Not verified for this market | Usage | Prioritized processing at a premium over the relevant Standard rate; Supported models and regions only |
| Flex | Not verified for this market | Usage | Discounted model rate with lower-priority synchronous processing; Supported models only; distinct from asynchronous Batch |
| Batch | Not verified for this market | Usage | Selected models: 50% below on-demand pricing according to AWS; Asynchronous processing of multiple inputs; separate model and region availability |
| Reserved | Not verified for this market | Usage | Reserved tokens-per-minute capacity for supported models; 1- or 3-month commitment; capacity-based price, billed monthly |
| Provisioned Throughput | Not verified for this market | Usage | Separately provisioned model capacity, billed by model unit and runtime; Prices and commitments differ from On-demand and Reserved offerings |
| GLM 5 (US: N. Virginia / Ohio / Oregon) | 1.00 / 3.20 USD | 1M input/output tokens, Standard | Example for one model and the named AWS Regions; First amount is input, second output; not a total Bedrock price |
| GLM 5 (Europe: Stockholm) | 1.20 / 3.84 USD | 1M input/output tokens, Standard | Example for one model and the named AWS Regions; First amount is input, second output; not a total Bedrock price |
| GLM 5.3 (Global cross-Region) | 1.68 / 5.28 USD | 1M input/output tokens, Standard | On-demand, Standard, Global cross-Region: 1.68 USD input and 5.28 USD output per 1M tokens. Cache read 0.312 USD. Cache write 30 minutes 2.10 USD.; US cross-Region: 1.848 USD input and 5.808 USD output. Cache read 0.3432 USD. Cache write 30 minutes 2.31 USD. Priority is 75 percent above Standard, Flex is 50 percent below. Eligible enterprise customers only. GA since 2026-10-05. This is the AWS price, not the direct Z.ai API. |

## Sources
All details regarding plans, limits, and features are based on official vendor documentation and the primary sources verified below.

- https://aws.amazon.com/bedrock/pricing/
- https://aws.amazon.com/bedrock/service-tiers/
- https://aws.amazon.com/about-aws/whats-new/2025/07/aws-free-tier-credits-month-free-plan/
- https://docs.aws.amazon.com/awsaccountbilling/latest/aboutv2/free-tier-FAQ.html
- https://docs.aws.amazon.com/bedrock/latest/userguide/cross-region-inference.html
- https://docs.aws.amazon.com/bedrock/latest/userguide/data-protection.html
- https://aws.amazon.com/about-aws/whats-new/2026/10/amazon-bedrock-glm-5-3/

---
*Source: AI Compare Ninja (2026) - https://www.aicompare.ninja/en/tool/amazon-bedrock*