Amazon EBS now provides visibility into the EBS volume initialization status for volumes created from EBS snapshots. You can use this status to determine when your volume becomes fully initialized when restoring from a snapshot and is fully ready to support latency-sensitive applications.

EBS volumes that are created from EBS snapshots undergo volume initialization, in which the storage blocks from the snapshot must be downloaded from Amazon S3 and written to the volume before you can access them. The volume initialization rate fluctuates throughout the initialization process, which could make completion times unpredictable. During initialization you may notice increased I/O latency and reduced performance. Our new volume initialization status feature allows you to validate when all blocks have been downloaded and written to the volume, enabling fully provisioned performance on your volume. Using this status, you can time your application launches to align with volume initialization completion. Monitor initialization progress in real-time and launch your applications only when volumes are fully ready, ensuring optimal performance from the start. When creating volumes with Provisioned Rate for Volume Initialization, you will also be able to see the estimated completion time for your volume initialization.

Volume initialization status is accessible by default for all EBS volumes. It is available in all AWS Regions, including the AWS GovCloud (US) Regions and AWS China Regions. You can start using this feature today through the AWS Management Console, AWS Command Line Interface (CLI), or AWS SDKs. To learn more about the new volume initialization status and how to access it, please visit the EBS initialize volume documentation. 

 

​Amazon EBS now provides visibility into the EBS volume initialization status for volumes created from EBS snapshots. You can use this status to determine when your volume becomes fully initialized when restoring from a snapshot and is fully ready to support latency-sensitive applications. EBS volumes that are created from EBS snapshots undergo volume initialization, in which the storage blocks from the snapshot must be downloaded from Amazon S3 and written to the volume before you can access them. The volume initialization rate fluctuates throughout the initialization process, which could make completion times unpredictable. During initialization you may notice increased I/O latency and reduced performance. Our new volume initialization status feature allows you to validate when all blocks have been downloaded and written to the volume, enabling fully provisioned performance on your volume. Using this status, you can time your application launches to align with volume initialization completion. Monitor initialization progress in real-time and launch your applications only when volumes are fully ready, ensuring optimal performance from the start. When creating volumes with Provisioned Rate for Volume Initialization, you will also be able to see the estimated completion time for your volume initialization. Volume initialization status is accessible by default for all EBS volumes. It is available in all AWS Regions, including the AWS GovCloud (US) Regions and AWS China Regions. You can start using this feature today through the AWS Management Console, AWS Command Line Interface (CLI), or AWS SDKs. To learn more about the new volume initialization status and how to access it, please visit the EBS initialize volume documentation.   

Amazon Nova Sonic—a speech-to-speech foundation model—now supports French, Italian, and German, expanding on its existing coverage of English and Spanish. This update includes six additional expressive voices, offering both masculine and feminine-sounding options, to help developers create more natural and inclusive conversational AI experiences across a wider range of languages.

In addition, Amazon Nova Sonic now integrates with LiveKit, an open-source WebRTC platform, and Pipecat, an open-source framework for building voice and multimodal AI agents. These integrations simplify the development of low-latency, real-time voice applications by removing the need to manage complex audio pipelines and streaming infrastructure. As an added capability, Nova Sonic now also supports integrations with Vonage and Twilio, extending deployment flexibility for telephony and communications use cases.

Amazon Nova Sonic is a speech-to-speech foundation model that delivers real-time, human-like voice conversations with low latency. Available in Amazon Bedrock via the bidirectional streaming API, the model understands streaming speech in various speaking styles and generates expressive speech responses that dynamically adapt to the prosody of input speech.

Amazon Nova Sonic is now available globally on Amazon Bedrock in three AWS Region. To learn more, read the AWS News Blog, Amazon Nova Sonic product page, and User Guide. To get started, visit the Amazon Bedrock Console.

 

​Amazon Nova Sonic—a speech-to-speech foundation model—now supports French, Italian, and German, expanding on its existing coverage of English and Spanish. This update includes six additional expressive voices, offering both masculine and feminine-sounding options, to help developers create more natural and inclusive conversational AI experiences across a wider range of languages. In addition, Amazon Nova Sonic now integrates with LiveKit, an open-source WebRTC platform, and Pipecat, an open-source framework for building voice and multimodal AI agents. These integrations simplify the development of low-latency, real-time voice applications by removing the need to manage complex audio pipelines and streaming infrastructure. As an added capability, Nova Sonic now also supports integrations with Vonage and Twilio, extending deployment flexibility for telephony and communications use cases. Amazon Nova Sonic is a speech-to-speech foundation model that delivers real-time, human-like voice conversations with low latency. Available in Amazon Bedrock via the bidirectional streaming API, the model understands streaming speech in various speaking styles and generates expressive speech responses that dynamically adapt to the prosody of input speech. Amazon Nova Sonic is now available globally on Amazon Bedrock in three AWS Region. To learn more, read the AWS News Blog, Amazon Nova Sonic product page, and User Guide. To get started, visit the Amazon Bedrock Console.  

Today, AWS introduces the AWS AI League, a program that helps organizations upskill their workforce by combining a fun competition with hands-on learning using AWS AI services such as Amazon SageMaker AI and Amazon Bedrock. The program offers a unique opportunity for both enterprises and developers to gain valuable and practical skills in fine-tuning, model customization and prompt engineering. Enterprises can apply to receive AWS credits to host internal AWS AI League competitions, fostering a culture of innovation within their organizations. Individual developers can also participate in the AWS AI League at select AWS Summits and AWS re:Invent, giving them a chance to compete while engaging with cutting-edge AI technologies and gaining skills crucial for developing advanced AI solutions.

AWS is committing up to $2 million in AWS credits and a championship prize pool of $25,000 to reward top performers at AWS re:Invent 2025. This significant investment underscores AWS commitment to advancing AI skills across the workforce and accelerating innovation in the field of generative AI.

For more information about the AWS AI League and how to participate, please visit the AWS AI League page. 

 

​Today, AWS introduces the AWS AI League, a program that helps organizations upskill their workforce by combining a fun competition with hands-on learning using AWS AI services such as Amazon SageMaker AI and Amazon Bedrock. The program offers a unique opportunity for both enterprises and developers to gain valuable and practical skills in fine-tuning, model customization and prompt engineering. Enterprises can apply to receive AWS credits to host internal AWS AI League competitions, fostering a culture of innovation within their organizations. Individual developers can also participate in the AWS AI League at select AWS Summits and AWS re:Invent, giving them a chance to compete while engaging with cutting-edge AI technologies and gaining skills crucial for developing advanced AI solutions. AWS is committing up to $2 million in AWS credits and a championship prize pool of $25,000 to reward top performers at AWS re:Invent 2025. This significant investment underscores AWS commitment to advancing AI skills across the workforce and accelerating innovation in the field of generative AI. For more information about the AWS AI League and how to participate, please visit the AWS AI League page.   

Starting today, customers can use the on-demand deployment option in Amazon Bedrock for Nova models that have been fine-tuned or distilled in Bedrock, or customized in SageMaker AI. Models customized on or after 7/16/2025 will be eligible.

This enables Bedrock customers to reduce costs by processing requests in real-time without requiring pre-provisioned compute resources. Customers only pay for what they use, helping reduce the need for an always on infrastructure.

Amazon Bedrock is a fully managed service that offers a choice of high-performing foundation models from leading AI companies via a single API. Amazon Bedrock also provides a broad set of capabilities customers need to build generative AI applications with security, privacy, and responsible AI built in.

Learn more in documentation here and here.

 

​Starting today, customers can use the on-demand deployment option in Amazon Bedrock for Nova models that have been fine-tuned or distilled in Bedrock, or customized in SageMaker AI. Models customized on or after 7/16/2025 will be eligible. This enables Bedrock customers to reduce costs by processing requests in real-time without requiring pre-provisioned compute resources. Customers only pay for what they use, helping reduce the need for an always on infrastructure. Amazon Bedrock is a fully managed service that offers a choice of high-performing foundation models from leading AI companies via a single API. Amazon Bedrock also provides a broad set of capabilities customers need to build generative AI applications with security, privacy, and responsible AI built in. Learn more in documentation here and here.  

Amazon CloudWatch now helps you observe generative AI applications and workloads, including agents deployed and operated with Amazon Bedrock AgentCore (Preview), providing insights into AI performance, health, and accuracy. You get an out-of-the-box view into latency, usage, and errors of your AI workloads to detect issues faster in components like model invocations and agents. You can also find issues faster using end-to-end prompt tracing of components like knowledge bases, tools and models. This feature is compatible with popular generative AI orchestration frameworks such as Strands Agents, LangChain, and LangGraph, offering flexibility with your choice of framework.

With this new feature, Amazon CloudWatch analyzes telemetry data across components of a generative AI application, helping quickly identify the source of errors. For example, you can pinpoint the source of inaccurate responses — whether from gaps in your VectorDB or incomplete RAG system retrials — using end-to-end prompt tracing, curated metrics and logs. This connected view of component interactions helps developers optimize workloads faster to deliver high levels of availability, accuracy, reliability, and quality. Developers can keep AI agents running smoothly by monitoring and assessing the fleet of agents in one place. The agent-curated view is available in the “AgentCore” tab in the CloudWatch console for genAI observability. Generative AI observability is integrated with other CloudWatch capabilities such as Application Signals, Alarms, Dashboards, Sensitive Data Protection, and Logs Insights, helping you seamlessly extend existing observability tools to monitor generative AI workloads.

This feature is available in Preview in 4 regions: US East (N.Virginia), US West (Oregon), Europe (Frankfurt), and Asia Pacific (Sydney). To learn more, visit documentation. CloudWatch pricing applies for collected and stored telemetry data.

 

​Amazon CloudWatch now helps you observe generative AI applications and workloads, including agents deployed and operated with Amazon Bedrock AgentCore (Preview), providing insights into AI performance, health, and accuracy. You get an out-of-the-box view into latency, usage, and errors of your AI workloads to detect issues faster in components like model invocations and agents. You can also find issues faster using end-to-end prompt tracing of components like knowledge bases, tools and models. This feature is compatible with popular generative AI orchestration frameworks such as Strands Agents, LangChain, and LangGraph, offering flexibility with your choice of framework. With this new feature, Amazon CloudWatch analyzes telemetry data across components of a generative AI application, helping quickly identify the source of errors. For example, you can pinpoint the source of inaccurate responses — whether from gaps in your VectorDB or incomplete RAG system retrials — using end-to-end prompt tracing, curated metrics and logs. This connected view of component interactions helps developers optimize workloads faster to deliver high levels of availability, accuracy, reliability, and quality. Developers can keep AI agents running smoothly by monitoring and assessing the fleet of agents in one place. The agent-curated view is available in the “AgentCore” tab in the CloudWatch console for genAI observability. Generative AI observability is integrated with other CloudWatch capabilities such as Application Signals, Alarms, Dashboards, Sensitive Data Protection, and Logs Insights, helping you seamlessly extend existing observability tools to monitor generative AI workloads. This feature is available in Preview in 4 regions: US East (N.Virginia), US West (Oregon), Europe (Frankfurt), and Asia Pacific (Sydney). To learn more, visit documentation. CloudWatch pricing applies for collected and stored telemetry data.  

Amazon Bedrock AgentCore enables developers to deploy and operate AI agents with the scale, reliability, and security critical to real-world applications. It provides purpose-built infrastructure to scale agents securely, powerful tools to enhance agent capabilities, and essential controls to ensure trustworthy operations. AgentCore services are modular and composable, allowing them to be used together or independently. They work with any model—in or outside of Amazon Bedrock—and any open-source agent framework, eliminating the trade-off between open-source flexibility and enterprise-grade security.

Amazon Bedrock AgentCore include services and tools that address the barriers to moving agents from proof of concept to production. AgentCore Runtime provides complete session isolation with low latency and supports long-running workloads up to 8 hours. AgentCore Memory enables agents to maintain both short-term and long-term memory across interactions with zero infrastructure management. AgentCore Gateway simplifies tool integration and discoverability, enabling developers to convert existing APIs and services into Model Context Protocol (MCP)-compatible tools with minimal code. AgentCore Browser Tool provides a secure, cloud-based browser runtime so agents can interact with web-based services and perform complex web tasks. AgentCore Code Interpreter offers a secure, sandbox environment so agents can execute code across multiple languages. AgentCore Observability provides real-time visibility into end-to-end agent execution and key operational metrics through dashboards powered by Amazon CloudWatch and is OpenTelemetry compatible. AgentCore Identity allows users to invoke agents by integrating with existing identity providers such as Amazon Cognito, Microsoft Entra ID, and Okta and enables agents to then securely access AWS resources and third-party tools and services.

The preview of Amazon Bedrock AgentCore is currently available in US East (N. Virginia), US West (Oregon), Asia Pacific (Sydney), and Europe (Frankfurt). Learn more about Amazon Bedrock AgentCore and it’s services in the News Blog and explore in-depth implementation details in the AgentCore documentation. For pricing information, visit the Amazon Bedrock AgentCore Pricing.

 

​Amazon Bedrock AgentCore enables developers to deploy and operate AI agents with the scale, reliability, and security critical to real-world applications. It provides purpose-built infrastructure to scale agents securely, powerful tools to enhance agent capabilities, and essential controls to ensure trustworthy operations. AgentCore services are modular and composable, allowing them to be used together or independently. They work with any model—in or outside of Amazon Bedrock—and any open-source agent framework, eliminating the trade-off between open-source flexibility and enterprise-grade security.
Amazon Bedrock AgentCore include services and tools that address the barriers to moving agents from proof of concept to production. AgentCore Runtime provides complete session isolation with low latency and supports long-running workloads up to 8 hours. AgentCore Memory enables agents to maintain both short-term and long-term memory across interactions with zero infrastructure management. AgentCore Gateway simplifies tool integration and discoverability, enabling developers to convert existing APIs and services into Model Context Protocol (MCP)-compatible tools with minimal code. AgentCore Browser Tool provides a secure, cloud-based browser runtime so agents can interact with web-based services and perform complex web tasks. AgentCore Code Interpreter offers a secure, sandbox environment so agents can execute code across multiple languages. AgentCore Observability provides real-time visibility into end-to-end agent execution and key operational metrics through dashboards powered by Amazon CloudWatch and is OpenTelemetry compatible. AgentCore Identity allows users to invoke agents by integrating with existing identity providers such as Amazon Cognito, Microsoft Entra ID, and Okta and enables agents to then securely access AWS resources and third-party tools and services.
The preview of Amazon Bedrock AgentCore is currently available in US East (N. Virginia), US West (Oregon), Asia Pacific (Sydney), and Europe (Frankfurt). Learn more about Amazon Bedrock AgentCore and it’s services in the News Blog and explore in-depth implementation details in the AgentCore documentation. For pricing information, visit the Amazon Bedrock AgentCore Pricing.  

AWS Marketplace now offers AI agents and tools from AWS Partners, allowing customers to find and buy third-party AI agent solutions with streamlined procurement and multiple deployment options. Customers can accelerate their discovery of AI agents and agent tools in a centralized catalog while enjoying the benefits of purchasing through AWS Marketplace and Partners can quickly bring their AI agent solutions to market.

Customers can explore AI agent products on the new “AI Agent & Tools” solution page. Using natural language, customers can search and receive results that match their specific use cases. When evaluating solutions, customers can review listings that support model context protocol (MCP) and agent-to-agent (A2A) standard protocols, along with various deployment options, to determine the best-fit solution for their needs. Customers can then purchase and deploy their chosen solutions through various paths, including Amazon Bedrock AgentCore Runtime, or add tools to AgentCore Gateway to accelerate agent development.

For AWS Partners, AI Agents and Tools in AWS Marketplace accelerates customer reach and adoption for agentic solutions. By listing their AI agents and tools, Partners can leverage established AWS Marketplace channels to streamline sales, offer flexible pricing, and provide secure AWS deployment options. Partners can categorize their offerings and highlight MCP and A2A protocol support, enhancing discoverability through advanced search and filtering in the AWS Marketplace catalog. Integration with Amazon Bedrock AgentCore services further simplifies deployment for customers, reducing time to value and providing a secure, scalable environment for customers building innovative agentic solutions.

Start exploring AI agent solutions in AWS Marketplace. Learn how AWS Partners can start selling by accessing the AWS Marketplace Seller Guide. 

 

​AWS Marketplace now offers AI agents and tools from AWS Partners, allowing customers to find and buy third-party AI agent solutions with streamlined procurement and multiple deployment options. Customers can accelerate their discovery of AI agents and agent tools in a centralized catalog while enjoying the benefits of purchasing through AWS Marketplace and Partners can quickly bring their AI agent solutions to market.
Customers can explore AI agent products on the new “AI Agent & Tools” solution page. Using natural language, customers can search and receive results that match their specific use cases. When evaluating solutions, customers can review listings that support model context protocol (MCP) and agent-to-agent (A2A) standard protocols, along with various deployment options, to determine the best-fit solution for their needs. Customers can then purchase and deploy their chosen solutions through various paths, including Amazon Bedrock AgentCore Runtime, or add tools to AgentCore Gateway to accelerate agent development. For AWS Partners, AI Agents and Tools in AWS Marketplace accelerates customer reach and adoption for agentic solutions. By listing their AI agents and tools, Partners can leverage established AWS Marketplace channels to streamline sales, offer flexible pricing, and provide secure AWS deployment options. Partners can categorize their offerings and highlight MCP and A2A protocol support, enhancing discoverability through advanced search and filtering in the AWS Marketplace catalog. Integration with Amazon Bedrock AgentCore services further simplifies deployment for customers, reducing time to value and providing a secure, scalable environment for customers building innovative agentic solutions. Start exploring AI agent solutions in AWS Marketplace. Learn how AWS Partners can start selling by accessing the AWS Marketplace Seller Guide.   

Today, Amazon Nova is introducing the most comprehensive suite of model customization capabilities made available for any proprietary model family. Available as ready-to-use recipes on SageMaker AI, these capabilities allow customers to adapt Nova Micro, Nova Lite, and Nova Pro across the model training lifecycle, including pre-training, supervised fine-tuning, and alignment.

Using these customization techniques, you can adapt Nova models to accurately reflect your proprietary knowledge, workflows, and brand in your generative AI applications while maintaining Nova’s industry-leading price performance and low latency. The techniques include Continued Pre-Training, Supervised Fine-Tuning (SFT), Direct Preference Optimization (DPO), Proximal Policy Optimization, and Knowledge Distillation — with support for both parameter-efficient and full-model training options across SFT, DPO and Distillation.

Nova customization recipes are available in SageMaker training jobs and SageMaker HyperPod, giving you flexibility to select the environment that best fits your infrastructure and scale requirements. You can deploy your customized models on Amazon Bedrock and invoke them via on-demand inference or Provisioned Throughput. On-demand inference is available only with parameter efficient training techniques.

Recipes for Amazon Nova on Amazon SageMaker AI are available in US East (N. Virginia).

To get started read Amazon Nova user guide and visit the GitHub repository to browse Nova specific SageMaker training recipes. 

 

​Today, Amazon Nova is introducing the most comprehensive suite of model customization capabilities made available for any proprietary model family. Available as ready-to-use recipes on SageMaker AI, these capabilities allow customers to adapt Nova Micro, Nova Lite, and Nova Pro across the model training lifecycle, including pre-training, supervised fine-tuning, and alignment. Using these customization techniques, you can adapt Nova models to accurately reflect your proprietary knowledge, workflows, and brand in your generative AI applications while maintaining Nova’s industry-leading price performance and low latency. The techniques include Continued Pre-Training, Supervised Fine-Tuning (SFT), Direct Preference Optimization (DPO), Proximal Policy Optimization, and Knowledge Distillation — with support for both parameter-efficient and full-model training options across SFT, DPO and Distillation. Nova customization recipes are available in SageMaker training jobs and SageMaker HyperPod, giving you flexibility to select the environment that best fits your infrastructure and scale requirements. You can deploy your customized models on Amazon Bedrock and invoke them via on-demand inference or Provisioned Throughput. On-demand inference is available only with parameter efficient training techniques. Recipes for Amazon Nova on Amazon SageMaker AI are available in US East (N. Virginia). To get started read Amazon Nova user guide and visit the GitHub repository to browse Nova specific SageMaker training recipes.   

Today, Amazon EKS announces support for up to 100,000 worker nodes in a cluster, enabling you to run ultra scale AI/ML training and inference workloads in a single cluster. With Amazon EC2’s new generation accelerated computing instance types, 100,000 worker nodes support up to 1.6 million Trainium chips with Trn2 instances and 800,000 NVIDIA GPUs with P5 and P6 instances in a single cluster. This enables you to run ultra scale AI/ML workloads that require all compute accelerators to be available within a single cluster, as these workloads cannot be easily distributed across multiple clusters.

The most advanced AI models with trillions of parameters demonstrate significantly enhanced capabilities in understanding context, reasoning, and solving complex tasks. To build and operate these increasingly powerful models, organizations require access to massive numbers of compute accelerators in a single cluster. Consolidated access to such a large pool of compute accelerators delivers crucial benefits: allows organizations to build and deploy more powerful AI models than ever before, reduces costs by efficiently sharing compute accelerators between training and inference workloads, and enables seamless use of existing AI/ML tools and frameworks that are not designed to work across clusters.

To learn more, see the launch blog.

 

​Today, Amazon EKS announces support for up to 100,000 worker nodes in a cluster, enabling you to run ultra scale AI/ML training and inference workloads in a single cluster. With Amazon EC2’s new generation accelerated computing instance types, 100,000 worker nodes support up to 1.6 million Trainium chips with Trn2 instances and 800,000 NVIDIA GPUs with P5 and P6 instances in a single cluster. This enables you to run ultra scale AI/ML workloads that require all compute accelerators to be available within a single cluster, as these workloads cannot be easily distributed across multiple clusters. The most advanced AI models with trillions of parameters demonstrate significantly enhanced capabilities in understanding context, reasoning, and solving complex tasks. To build and operate these increasingly powerful models, organizations require access to massive numbers of compute accelerators in a single cluster. Consolidated access to such a large pool of compute accelerators delivers crucial benefits: allows organizations to build and deploy more powerful AI models than ever before, reduces costs by efficiently sharing compute accelerators between training and inference workloads, and enables seamless use of existing AI/ML tools and frameworks that are not designed to work across clusters. To learn more, see the launch blog.  

Amazon S3 Vectors delivers purpose-built, cost-optimized vector storage for AI agents, AI inference, and semantic search of your content stored in Amazon S3. By reducing the cost of uploading, storing, and querying vectors by up to 90%, S3 Vectors makes it cost-effective to create and use large vector datasets to improve the memory and context of AI agents as well as semantic search results of your S3 data. Designed to provide the same elasticity, scale, and durability as Amazon S3, S3 Vectors lets you store and search data with sub-second query performance. It’s ideal for applications that need to build and maintain vector indexes so you can organize and search through massive amounts of information. S3 Vectors provides a simple and flexible API for operations such as finding similar scenes in petabyte-scale video archives, identifying collections of related business documents, or detecting rare patterns in diagnostic collections including millions of medical images.

S3 Vectors is natively integrated with Amazon Bedrock Knowledge Bases so that you can reduce the cost of using large vector datasets for retrieval-augmented generation (RAG). You can also use S3 Vectors with Amazon OpenSearch Service to lower storage costs for infrequent queried vectors, and then quickly move them to OpenSearch as demands increase or to enhance search capabilities.

S3 Vectors introduces a new bucket type optimized for durable, low-cost vector storage. It includes a dedicated set of APIs for you to store, access, and query vectors without provisioning any infrastructure. Within a vector bucket, you can organize vector data within vector indexes and elastically scale up to 10,000 indexes per bucket. When creating a Knowledge Base in Amazon Bedrock or Amazon SageMaker Unified Studio, you can select an S3 vector index as your vector store or use the Quick Create workflow to set one up. Within OpenSearch, you can adopt a tiered strategy to store large vector datasets in S3 for near real-time access while effortlessly activating the vector data with the highest performance requirements in OpenSearch.

Amazon S3 Vectors preview is now available in the US East (N. Virginia), US East (Ohio), US West (Oregon), Asia Pacific (Sydney), and Europe (Frankfurt) Regions. To learn more, visit the product page, S3 pricing page, documentation, and AWS News Blog. 

 

​Amazon S3 Vectors delivers purpose-built, cost-optimized vector storage for AI agents, AI inference, and semantic search of your content stored in Amazon S3. By reducing the cost of uploading, storing, and querying vectors by up to 90%, S3 Vectors makes it cost-effective to create and use large vector datasets to improve the memory and context of AI agents as well as semantic search results of your S3 data. Designed to provide the same elasticity, scale, and durability as Amazon S3, S3 Vectors lets you store and search data with sub-second query performance. It’s ideal for applications that need to build and maintain vector indexes so you can organize and search through massive amounts of information. S3 Vectors provides a simple and flexible API for operations such as finding similar scenes in petabyte-scale video archives, identifying collections of related business documents, or detecting rare patterns in diagnostic collections including millions of medical images.
S3 Vectors is natively integrated with Amazon Bedrock Knowledge Bases so that you can reduce the cost of using large vector datasets for retrieval-augmented generation (RAG). You can also use S3 Vectors with Amazon OpenSearch Service to lower storage costs for infrequent queried vectors, and then quickly move them to OpenSearch as demands increase or to enhance search capabilities.
S3 Vectors introduces a new bucket type optimized for durable, low-cost vector storage. It includes a dedicated set of APIs for you to store, access, and query vectors without provisioning any infrastructure. Within a vector bucket, you can organize vector data within vector indexes and elastically scale up to 10,000 indexes per bucket. When creating a Knowledge Base in Amazon Bedrock or Amazon SageMaker Unified Studio, you can select an S3 vector index as your vector store or use the Quick Create workflow to set one up. Within OpenSearch, you can adopt a tiered strategy to store large vector datasets in S3 for near real-time access while effortlessly activating the vector data with the highest performance requirements in OpenSearch. Amazon S3 Vectors preview is now available in the US East (N. Virginia), US East (Ohio), US West (Oregon), Asia Pacific (Sydney), and Europe (Frankfurt) Regions. To learn more, visit the product page, S3 pricing page, documentation, and AWS News Blog.