Amazon Bedrock Connector

Content Type: Module
Categories: Connectors,AWS,Artificial Intelligence

Overview

Amazon Bedrock is a fully managed service that offers a choice of high-performing foundation models (FMs) from leading AI companies like AI21 Labs, Anthropic, Cohere, Meta, Stability AI, Mistral AI and Amazon via a single API, along with a broad set of capabilities you need to build generative AI applications, simplifying development while maintaining privacy and security.

The Amazon Bedrock Connector allows you to integrate with Amazon Bedrock.

As of version 3.0.0 the connector depends on the GenAICommons module and provides the following out-of-the box implementations:

  • ChatCompletions (without history)
  • ChatCompletions (with history)
  • Retrieve and Generate.

 

Thanks to these implementations, it is possible to use Claude and Titan models as assistants in both the GenAI Showcase app as well as the AI Bot Starter app out of the box. Please have a look into the GenAICommons documentation to find more information about how to build the required request structure and connection object to reuse the operations. 

 

In addition, the connector provides the following actions:

  • List Foundation Models
  • Invoke Model (generic)
  • Retrieve
  • StartIngestionJob and GetIngestionJob (make it possible to sync changes to KnowledgeBases)
  • List knowledge bases
  • Invoke Agent

 

For even more extensive examples, please find the Amazon Bedrock Showcase App on the marketplace to learn and get some inspiration on how the Amazon Bedrock Connector can be used in your Mendix application and to see additional example microflows for invoking different foundation models available on Amazon Bedrock. 

For more information about our AWS Connector strategy, please see the Mendix Evaluation Guide.

Documentation

Please see the documentation on Mendix Docs

Releases

Version: 3.0.0
Framework Version: 9.24.2
Release Notes: New Features: - two ChatCompletions operations which provide reusable logic to use all Claude (+Vision) and Titan text generation models as assistants. Improvements: - upgrade to 9.24.2. - added dependency to GenAICommons module. - a RetrieveAndGenerate operation which can be used with Claude v2.1 and Claude v3 Sonnet. - a toolbox with logic to build a GenAiCommons.Request and GenAiCommons.Connection.
Version: 2.3.0
Framework Version: 9.18.0
Release Notes: - InvokeModel examples for specific models replaced with Claude 3 and Titan Image models. - Added example for a secure published REST service to use for an Amazon Bedrock agent - Fixed bug where blank session ID threw an error when using RetrieveAndGenerate
Version: 2.2.0
Framework Version: 9.18.0
Release Notes: - Invoke Agent operation has been added - Invoke Model examples are now usable in every AWS region
Version: 2.1.3
Framework Version: 9.18.0
Release Notes: - Small bug fix for the RetrieveAndGenerate java action
Version: 2.1.2
Framework Version: 9.18.0
Release Notes: - The ListFoundationModels action has been extended to include more features in its response. - Two examples of how to invoke a specific model using the InvokeGenericModel action were added to the connector to help users utilize the Amazon Bedrock Connector. - Minor change to Domain Model for readability.
Version: 2.1.0
Framework Version: 9.18.0
Release Notes: New java actions added: - ListFoundationModels - InvokeModel - Retrieve - RetrieveAndGenerate - StartIngestionJob and GetIngestionJob (make it possible to sync changes to KnowledgeBases) - ListKnowledgeBases
Version: 2.0.0
Framework Version: 9.18.0
Release Notes: - Connector updated to work with Authentication Connector 3.0 and newer versions
Version: 1.0.3
Framework Version: 9.18.0
Release Notes: - The bug that was giving the 400 error on invoking foundational models has now been resolved.
Version: 1.0.1
Framework Version: 9.18.0
Release Notes: Improvements - We added a constant, Timeout, so that the time out can be configured for API calls
Version: 1.0.0
Framework Version: 9.18.0
Release Notes: Initial release