Amazon Bedrock
AWS service that serves models from several providers behind one API, with tooling for agents and evaluation.
Find alternatives to Amazon Bedrock
What it does
- Model inference
- Agent orchestration
- Evaluation and observability
Features 15
- Knowledge BasesGrounds model responses and agent decisions in organizational documents and connected enterprise information
- Smart ParsingAutomatically selects parsing strategies for diverse data types while preserving useful document context
- Agentic RetrievalReasons across enterprise sources and iterates retrieval steps for complex information requests
- GuardrailsApplies configurable content safeguards, sensitive information filters, and hallucination checks across models
- Automated Reasoning checksValidates generated answers against logical rules and identifies errors, corrections, and assumptions
- Managed Agents, powered by OpenAICombines OpenAI models and the Codex harness with AWS infrastructure for persistent agents
- Prompt cachingReuses repeated prompt prefixes to avoid recomputation across supported model requests
- Prompt ManagementCreates, evaluates, versions, optimizes, and runs prompts for foundation model applications
- Intelligent Prompt RoutingRoutes prompts between models within a family to balance response quality and cost
- Model DistillationTransfers capabilities from larger teacher models into smaller models using generated training responses
- Amazon Bedrock FlowsVisually connects prompts, agents, knowledge bases, and services into testable generative AI workflows
- Amazon Bedrock Managed AgentsRuns OpenAI-powered agents with managed inference, memory, skills, and AWS governance controls
- Supervised fine-tuningTrains foundation models on proprietary examples to improve task performance and domain-specific behavior
- Reinforcement fine-tuningImproves model responses through feedback that teaches quality standards and user preferences
- EvaluationsEvaluates model and retrieval performance using automated metrics, judge models, and human feedback
How it compares
-
Generative AI Capabilities
Our enterprise cloud, software and hardware offerings compete directly with certain offerings from some of the largest and most competitive companies in the world, including Adobe Systems Incorporated, Alphabet Inc., Amazon.com, Inc., Cisco Systems, Inc., Intel Corporation, International Business Machines Corporation, Microsoft Corporation, Salesforce, Inc. and SAP SE, as well as other companies like Hewlett-Packard Enterprise and Workday, Inc.
Regulatory filing · 18 Sep 2026 -
Azure OpenAI in Foundry Models
Those actors would be the 'Big 3' cloud giants: Amazon Web Services, Microsoft and Google, trailed by IBM.
Research report · 19 Sep 2026 -
OpenRouter
Given that AWS has already launched a native intelligent prompt routing feature that appears competitive with OpenRouter's, hyperscalers act as both supply-side partners on OpenRouter and potential competitors in model routing.
Research report · 19 Sep 2026 -
Uniphore Business AI Cloud
AWS Bedrock is infrastructure for ML teams. Uniphore BAIC is a complete enterprise AI platform — data fabric, domain models, process discovery, and governed agents — all four layers, ready to deploy.
Official documentation · 19 Sep 2026 -
Tinker
AWS Bedrock, Microsoft Azure AI, and Google Vertex AI provide model-agnostic platforms that let enterprises access multiple foundation models through unified APIs. These platforms compete directly with Thinking Machines' customization layer.
Research report · 19 Sep 2026 -
Elastic Agent Builder
General-purpose frameworks like LangGraph and Amazon Bedrock Agents leave retrieval for you to add and configure.
Official documentation · 13 Oct 2026
Sources
- Pricing
-
Published per-token, per-hour and per-month rates; no free tier is named on the pricing page.
- Constraints
-
Official documentation · 2 Oct 2026
Derived from the security facts on this page, from the same source.
- Security
-
Official documentation · 2 Oct 2026
Bedrock-specific official FAQ page; statements quoted are explicitly about Amazon Bedrock (not just generic AWS), including the no-training and no-sharing-with-model-providers claims. The certifier's own register adds what the page above did not state. Auditor listing: ISO/IEC 42001 -- Schellman certificate register https://integrator.schellman.com/public/v2/certificate?search=Amazon%20Web%20Services%2C%20Inc.: 'Amazon Web Services, Inc. / ISO/IEC 42001:2023 / 1821017-2 / 2027-11-14'; scope names 'Amazon Bedrock'.
- Sold within
-
AWS publishes a rate card for this service on its own pricing page, read 2026-09-06: it is metered on its own and appears as its own line on the bill. That is the ceiling rule's own test for a standalone purchase.
- Interfaces
-
Official documentation · 29 Sep 2026
Docs give REST-style Converse/Invoke/Messages/Responses API calls and a boto3 (AWS's own Python SDK) 'bedrock-runtime' client example: 'client = boto3.client("bedrock-runtime", ...); client.converse(...)'.
- Intended for
-
enterprise: 'startups to global enterprises across every industry', '100,000+ organizations', 'enterprise security', 'production scale'; developer: sections on customization, agent frameworks and model choice; public-sector: FedRAMP High and HIPAA-eligible compliance called out for regulated sectors.
Also from Amazon
-
Amazon Kendra Data service
A managed AWS enterprise search service that answers natural-language queries and powers retrieval-augmented generation over indexed content.
-
Amazon Lex API service
A managed AWS service for building voice and text conversational bots, built on the same technology as Alexa.
-
Amazon Nova Act Agent platform
An AWS platform for building and running AI agents that automate browser-based UI workflows, billed per agent-hour of active work.
-
Amazon Polly API service
A managed AWS API that converts text into synthesized speech across more than 40 languages and voice styles.
-
Amazon Q Developer Developer tool
An AWS coding assistant that generates, reviews, tests and autonomously modifies code inside IDEs and the CLI; AWS will discontinue support for the IDE plugins on 2027-04-30 and points to Kiro.
-
Amazon Quick Agent platform
AWS's subscription AI work assistant combining chat-based research, business-intelligence dashboards and workflow automation, positioned as the successor to Amazon Q Business; also turns a conversation into a doc, spreadsheet, branded deck, image or dashboard.
-
Amazon Rekognition API service
A managed AWS API for image and video analysis, including object and face detection, text extraction from images and content moderation.
-
Amazon SageMaker AI Platform
AWS's unified platform for building, training and deploying machine learning and generative AI models.
-
Amazon Textract API service
A managed AWS API that extracts text, handwriting, forms and tables from scanned documents.
-
Amazon Transcribe API service
A managed AWS API that converts speech to text using an automatic speech recognition model.
-
AWS Inferentia Hardware
AWS's custom silicon for machine learning inference, purchased via Amazon EC2 Inf1 and Inf2 instances rather than as a standalone chip.
-
AWS Trainium Hardware
Amazon's own training accelerator, rented by the hour inside EC2 instances rather than sold as a part.
-
Alexa+ Application
Amazon's generative-AI Alexa: a conversational assistant across Echo devices and the Alexa app that handles multi-step requests, remembers context and acts on behalf of the user.
-
Amazon Comprehend API service
Natural-language-processing service that extracts key phrases, sentiment, entities, document classification and PII redaction from text.
-
Amazon Connect Customer Platform
Cloud contact-centre service, formerly simply 'Amazon Connect', now positioned as an agentic-AI customer-experience solution: AI agents and intelligent routing across voice, chat, email, SMS, web and WhatsApp, designed on a no-code canvas, with real-time performance tracking for agents and supervisors.
-
Amazon Connect Health Agent platform
Agentic AI for healthcare providers and health-tech developers: patient verification, appointment management, ambient clinical documentation, patient insights and medical coding, delivered inside Amazon Connect Customer for patient engagement and via an SDK at the point of care.
-
Kiro Developer tool
Agentic AI development platform (IDE, CLI, web, mobile) that turns prompts into executable specs, validates code with property-based testing and runs parallel agents across large codebases.
-
Amazon Personalize API service
Recommendation engine trained on user-interaction data to deliver real-time personalized recommendations across web, app and marketing channels.
-
Amazon Translate API service
Neural machine translation service for batch and real-time cross-lingual text translation with brand-terminology customization.
-
Amazon Bedrock AgentCore Platform
Production AI-agent operational platform (framework- and model-agnostic) providing security controls, tool integration, debugging and evaluation for deployed agents.
-
AWS HealthScribe API service
HIPAA-eligible service that transcribes patient-clinician conversations and generates preliminary clinical documentation with evidence mapping back to the transcript.
-
AWS Transform Platform
Agentic-AI enterprise IT transformation workbench that accelerates cloud migration, application modernization (mainframe, Windows, custom code) and continuous tech-debt reduction.
-
AWS Continuum Platform
A security platform that discovers, prioritizes, validates, and remediates software security risks.
-
AWS DevOps Agent AI agent
An agent that autonomously investigates operational incidents; also reviews code for release readiness and runs autonomous release testing.
-
Amazon Ads Agent Platform
An AI advertising platform that unifies omnichannel advertising and automates campaign execution and optimization.
-
Amazon Business Assistant Application
An assistant that provides conversational answers and guidance to procurement teams.
-
Amazon CodeGuru Profiler Developer tool
A developer tool that identifies expensive lines of application code through runtime analysis; also discovers code anomalies and troubleshoots performance issues through continuous monitoring.
-
Amazon Comprehend Medical API service
A natural language processing service that extracts health information from medical text.
-
Amazon DevOps Guru Application
An operations service that uses machine learning to detect abnormal application behavior.
-
Amazon Quick Automate Agent platform
An application that creates automations from natural language or documentation using specialized AI agents.
-
Amazon Quick Flows Application
A component that lets users build workflows to automate repetitive tasks.
-
Amazon SageMaker Autopilot Developer tool
An AutoML component that automatically creates and selects machine learning models.
-
Amazon SageMaker Canvas Application
An application for building and training machine learning models without writing code.
-
Amazon SageMaker Data Wrangler Developer tool
A component for preparing data, engineering features, and detecting data quality issues for machine learning.
-
Amazon SageMaker Feature Store Data service
A repository for storing, sharing, and managing features used by machine learning models.
-
Amazon SageMaker HyperPod Infrastructure service
Infrastructure for training, fine-tuning, serving, and observing models.
-
Amazon SageMaker JumpStart Developer tool
A machine learning hub for evaluating, comparing, and selecting foundation models.
-
Amazon SageMaker Pipelines Developer tool
A serverless service for orchestrating and monitoring repeatable machine learning workflows.
-
Amazon SageMaker Studio Developer tool
A web interface for developing, training, evaluating, and deploying AI models.
-
Amazon Bedrock Data Automation API service
Amazon Bedrock Data Automation uses generative AI to transform multimodal data into structured formats.
-
Amazon Connect Decisions Application
Amazon Connect Decisions provides supply chain planning and intelligence with AI teammates.
-
Amazon Connect Talent Application
Amazon Connect Talent uses AI teammates to assess and interview candidates.
-
Amazon Nova Forge Platform
Amazon Nova Forge is a service for building custom frontier models using Nova and proprietary training data.
-
Amazon SageMaker Ground Truth Data service
Amazon SageMaker Ground Truth creates training datasets and offers automated labeling for existing customers.
-
Amazon Quick Research AI agent
Research agent inside Amazon Quick: given a research objective it plans, searches the user's connected company data and the public web, and returns a customised report with citations to its sources.
-
Amazon Quick Sight Application
Business-intelligence part of Amazon Quick: dashboards, charts and embedded analytics over a company's data sources, with natural-language what-if analysis and agent-assisted exploration.
-
Alexa for Shopping Application
Amazon's AI shopping assistant in the Amazon Shopping app and on Amazon.com: answers shopping questions in the search bar, builds shopping guides and product comparisons, shows price history, and automates deal-finding and cart-building from the user's preferences.
-
Amazon Bedrock Guardrails API service
Configurable safeguards for generative AI applications: filters harmful text and image content, blocks denied topics and prompt attacks, redacts personal information and detects hallucinations, on both user inputs and model responses, for Bedrock or any other model via an API.
-
Amazon Bedrock Knowledge Bases API service
Managed retrieval for generative AI: connects a company's documents and data sources so foundation models and agents answer from that data, including a fully managed knowledge base and structured-data retrieval that generates SQL.
-
Amazon SageMaker Unified Studio Platform
Single browser-based environment for data and AI development: find governed data across an organisation, then query it with SQL, work in serverless notebooks with a built-in AI agent, train models and build generative AI applications in shared projects.
Something wrong here? Send a correction — quote this product id: aws-bedrock.