We are excited to announce the preview of Custom Model Import for Amazon Bedrock. Now you can import customized models into Amazon Bedrock to accelerate your generative AI application development. This new feature allows you to leverage your prior model customization investments within Amazon Bedrock and consume them in the same fully-managed manner as Bedrock’s existing models. For supported architectures such as Llama, Mistral, or Flan T5, you can now import models customized anywhere and access them on-demand.

Knowledge Bases for Amazon Bedrock is a fully managed Retrieval-Augmented Generation (RAG) capability that allows you to connect foundation models (FMs) to internal company data sources to deliver more relevant and accurate responses. Knowledge Bases now supports adding multiple data sources, across accounts.

Knowledge Bases for Amazon Bedrock allows you to connect foundation models (FMs) to internal company data sources to deliver more relevant, context-specific, and accurate responses. Knowledge Bases (KB) now provides a real-time, zero-setup, and low-cost method to securely chat with single documents. 

Amazon Titan Image Generator enables content creators with rapid ideation and iteration resulting in high efficiency image generation. The Amazon Titan Image Generator model is now generally available in Amazon Bedrock, helping you easily build and scale generative AI applications with new image generation and image editing capabilities.

Model Evaluation on Amazon Bedrock allows you to evaluate, compare, and select the best foundation models for your use case. Amazon Bedrock offers a choice of automatic evaluation and human evaluation. You can use automatic evaluation with predefined algorithms for metrics such as accuracy, robustness, and toxicity. Additionally, for those metrics or subjective and custom metrics, such as friendliness, style, and alignment to brand voice, you can set up a human evaluation workflow with a few clicks. Human evaluation workflows can leverage your own employees or an AWS-managed team as reviewers. Model evaluation provides built-in curated datasets or you can bring your own datasets.

Agents for Amazon Bedrock enable generative AI applications to automate multi-step tasks across company systems and data sources. Agents removes the undifferentiated lifting of orchestration, infrastructure hosting and management, and we’re making building Agents easier than ever.

You can now access Meta’s Llama 3 models, Llama 3 8B and Llama 3 70B, in Amazon Bedrock. Meta Llama 3 is designed for you to build, experiment, and responsibly scale your generative artificial intelligence applications. You can now use these two new Llama 3 models in Amazon Bedrock enabling you to easily experiment with and evaluate even more top foundation models for your use case.

Amazon Titan Image Generator’s new watermark detection feature is now generally available in Amazon Bedrock. All Amazon Titan-generated images contain an invisible watermark, by default. The watermark detection mechanism allows you to identify images generated by Amazon Titan Image Generator, a foundation model that allows users to create realistic, studio-quality images in large volumes and at low cost, using natural language prompts.

AWS Transfer Family customers can now use SFTP connectors to list files stored in remote SFTP servers, enabling visibility into the contents of directories in remote SFTP file systems and transfer files when file names are not known in advance.

Today, we’re excited to announce the availability of a new capability of Amazon Q to analyze issues for complexity and propose splitting the work into separate tasks.