Skip to main content

What is Bedrock Chat?

Bedrock Chat (BrChat) is an enterprise-ready, multilingual generative AI chat platform powered by Amazon Bedrock. It enables organizations to deploy a fully-featured AI chat application with custom bot creation, knowledge base integration, and task automation capabilities. Bedrock Chat Demo

Key Features

Bedrock Chat provides a comprehensive set of features for building production-grade AI applications:

Multi-Model Chat

Support for Claude, Amazon Nova, Llama, Mistral, and DeepSeek models via Amazon Bedrock API

Custom Bots with RAG

Create specialized bots with custom instructions and knowledge bases using Retrieval-Augmented Generation

Bot Store

Share and discover bots in a centralized marketplace with fine-grained permission controls

AI Agents

Automate complex tasks with ReAct agents that use tools and break down multi-step queries

API Publishing

Deploy custom bots as standalone REST APIs with dedicated endpoints and authentication

Admin Controls

Manage users, mark essential bots, analyze usage, and control API access

Bot Personalization

Create custom bots with your own instructions and knowledge using RAG (Retrieval-Augmented Generation). Bots can be shared among application users via the bot store marketplace and published as standalone APIs. Custom Bot Creation
For governance reasons, only users in the CreatingBotAllowed group can create custom bots. Configure this via Amazon Cognito User Pools console or AWS CLI.

Multi-Tenant Knowledge Bases

Amazon Bedrock Knowledge Bases has a default limit of 100 Knowledge Bases per AWS account. Bedrock Chat supports multi-tenant mode, where a shared Knowledge Base is used across multiple bots with metadata-based filtering using Bot IDs. Newly created bots automatically use multi-tenant mode. To migrate existing bots:
aws dynamodb execute-statement --statement "UPDATE \"$BotTableNameV3\" SET BedrockKnowledgeBase.type='shared' SET SyncStatus='QUEUED' WHERE PK='$UserID' AND SK='BOT#$BotID'"
aws stepfunctions start-execution --state-machine-arn $EmbeddingStateMachineArn

Administrative Features

Bedrock Chat provides comprehensive administrative controls:
  • API Management: Create and manage published bot APIs
  • Essential Bots: Mark important bots for organization-wide visibility
  • Usage Analytics: Analyze chat and bot usage patterns
  • User Management: Control permissions and group memberships
Bot Analytics Dashboard

Agent Functionality

By using the Agent functionality, your chatbot can automatically handle complex tasks. Agents can retrieve information from external tools and break down tasks into multiple steps for processing. Agent Functionality Learn more about agents in the documentation.

Supported Languages

Bedrock Chat supports 15+ languages with automatic detection:
  • English, Japanese (日本語), Korean (한국어), Chinese (中文)
  • French (Français), German (Deutsch), Spanish (Español), Italian
  • Norwegian (Norsk), Thai (ไทย), Indonesian, Malay, Vietnamese
  • Polish (Polski), Portuguese (Português Brasil)

Why Choose Bedrock Chat?

Built on AWS managed services with no infrastructure to manage. Automatically scales with your usage.
Multiple authentication providers, IP restrictions via AWS WAF, data isolation with Amazon Cognito, and AWS IAM integration.
All AI model interactions happen through Amazon Bedrock within your AWS account. No data leaves AWS.
Multi-tenant knowledge bases, configurable replicas, and Lambda SnapStart reduce operational costs.
Configure models, branding, authentication providers, security policies, and deployment parameters.

Getting Started

Quick Deployment

Deploy Bedrock Chat to your AWS account in under 40 minutes using CloudShell

Architecture Overview

Learn about the AWS services and architecture patterns used in Bedrock Chat

AWS Workshop

Follow our comprehensive hands-on workshop to learn all features

CDK Deployment

Advanced deployment with AWS CDK for custom configurations

Version 3.x Notice

V3 released. V3 is NOT compatible with V2 or earlier versions. To update, carefully review the migration guide. Without proper migration, BOTS FROM V2 WILL BECOME UNUSABLE.

Next Steps

1

Deploy Bedrock Chat

Follow our quickstart guide to deploy Bedrock Chat to your AWS account
2

Configure security

3

Create custom bots

4

Explore advanced features

Build docs developers (and LLMs) love