aws.amazon.com

Amazon Bedrock now supports Responses API on new OpenAI API-compatible service endpoints. Responses API enables developers to achieve asynchronous inference for long-running inference workloads, simplifies tool use integration for agentic workflows, and also supports stateful conversation management. Instead of requiring developers to pass the entire conversation history with each request, Responses API enables them to automatically rebuild context without manual history management. These new service endpoints support both streaming and non-streaming modes, enable reasoning effort support within Chat Completions API, and require only a base URL change for developers to integrate within existing codebases with OpenAI SDK compatibility. Chat Completions with reasoning effort support is available for all Amazon Bedrock models that are powered by Mantle, a new distributed inference engine for large-scale machine learning model serving on Amazon Bedrock. Mantle simplifies and ex

Starting today, new general purpose Amazon Elastic Compute Cloud (Amazon EC2) M9g instances, powered by AWS Graviton5 processors, are available in preview. AWS Graviton5 is the latest in the Graviton family of processors that are custom designed by AWS to provide the best price performance for workloads in Amazon EC2. These instances offer up to 25% better compute performance, and higher networking and Amazon Elastic Block Store (Amazon EBS) bandwidth than AWS Graviton4-based M8g instances. They are up to 30% faster for databases, up to 35% faster web applications, and up to 35% faster for machine learning workloads compared to M8g. M9g instances are built on the AWS Nitro System, a collection of hardware and software innovations designed by AWS. The AWS Nitro System enables the delivery of efficient, flexible, and secure cloud services with iso

Amazon SageMaker HyperPod now supports checkpointless training, a new foundational model training capability that mitigates the need for a checkpoint-based job-level restart for fault recovery. Checkpointless training maintains forward training momentum despite failures, reducing recovery time from hours to minutes. This represents a fundamental shift from traditional checkpoint-based recovery, where failures require pausing the entire training cluster, diagnosing issues manually, and restoring from saved checkpoints, a process that can leave expensive AI accelerators idle for hours, costing your organization wasted compute. Checkpointless training transforms this paradigm by preserving the model training state across the distributed cluster, automatically swapping out faulty training nodes on the fly and using peer-to-peer state transfer from healthy accelerators for failure recovery. By mitigating checkpoint dependencies during recovery, checkpointless training can he

In May, we open sourced the Strands Agents SDK, an open source python framework that takes a model-driven approach to building and running AI agents in just a few lines of code. Today, we’re announcing that TypeScript support is available in preview. Now, developers can choose between Python and TypeScript for building Strands Agents. TypeScript support in Strands has been designed to provide an idiomatic TypeScript experience with full type safety, async/await support, and modern JavaScript/TypeScript patterns. Strands can be easily run in client applications, in browsers, and server-side applications in runtimes like AWS Lambda and Bedrock AgentCore. Developers can also build their entire stack in Typescript using the AWS CDK. We’re also announcing three additional updates for the Strands SDK. First, edge device support for Strands Agents is generally available, extending the SDK with bidirectional streaming and additional local model providers like llama.cpp

Amazon Web Services (AWS) announces a new serverless model customization capability that empowers AI developers to quickly customize popular models with supervised fine-tuning and the latest techniques like reinforcement learning. Amazon SageMaker AI is a fully managed service that brings together a broad set of tools to enable high-performance, low-cost AI model development for any use case.  Many AI developers seek to customize models with proprietary data for improved accuracy, but this often requires lengthy iteration cycles. For example, AI developers must define a use case and prepare data, select a model and customization technique, train the model, then evaluate the model for deployment. Now AI developers can simplify the end-to-end model customization workflow, from data preparation to evaluation and deployment, and accelerate the process. With an easy-to-use interface, AI developers can quickly get started and customize popular models, incl

Amazon Bedrock now supports reinforcement fine-tuning, helping you improve model accuracy without needing deep machine learning expertise or large sums of labeled data. Amazon Bedrock automates the reinforcement fine-tuning workflow, making this advanced model customization technique accessible to everyday developers. Models learn to align with your specific requirements using a small set of prompts rather than the large sums of data needed for traditional fine-tuning methods, enabling teams to get started quickly. This capability teaches models through feedback on multiple possible responses to the same prompt, improving their judgement of what makes a good response. Reinforcement fine-tuning in Amazon Bedrock delivers 66% accuracy gains on average over base models so you can use smaller, faster, and more cost-effective model variants while maintaining high quality. Organizations struggle to adapt AI models to their unique business needs, forcing them to choose between gene

Amazon SageMaker HyperPod now supports elastic training, enabling organizations to accelerate foundation model training by automatically scaling training workloads based on resource availability and workload priorities. This represents a fundamental shift from training with a fixed set of resources, as it saves hours of engineering time spent reconfiguring training jobs based on compute availability. Any change in compute availability previously required manually halting training, reconfiguring training parameters, and restarting jobs—a process that requires distributed training expertise and leaves expensive AI accelerators sitting idle during training job reconfiguration. Elastic training automatically expands training jobs to absorb idle AI accelerators and seamlessly contracting when higher-priority workloads need resources—all without halting training entirely. By eliminating manual reconfiguration overhead and ensuring continuous utilization of available co

Starting today, new general purpose high-frequency high-network Amazon Elastic Compute Cloud (Amazon EC2) M8azn instances are available for preview. These instances are powered by fifth generation AMD EPYC (formerly code named Turin) processors, offering the highest maximum CPU frequency, 5GHz in the cloud. The M8azn instances offer up to 2x compute performance versus previous generation M5zn instances. These instances also deliver 24% higher performance than M8a instances. M8azn instances are built on the AWS Nitro System, a collection of hardware and software innovations designed by AWS. The AWS Nitro System enables the delivery of efficient, flexible, and secure cloud services with isolated multitenancy, private networking, and fast local storage. These instances are ideal for applications such as gaming, high-performance computing, high-frequency trading (HFT), CI/CD, and simulation modeling for the automotive, aerospace,

AWS announces the Apache Spark upgrade agent, a new capability that accelerates Apache Spark version upgrades for Amazon EMR on EC2 and EMR Serverless. The agent converts complex upgrade processes that typically take months into projects spanning weeks through automated code analysis and transformation. Organizations invest substantial engineering resources analyzing API changes, resolving conflicts, and validating applications during Spark upgrades. The agent introduces conversational interfaces where engineers express upgrade requirements in natural language, while maintaining full control over code modifications. The Apache Spark upgrade agent automatically identifies API changes and behavioral modifications across PySpark and Scala applications. Engineers can initiate upgrades directly from SageMaker Unified Studio, Kiro CLI or IDE of their choice with the help of MCP (Model Context Protocol) compatibility. During the upgrade process, the agent analyzes existing code an

Today, Amazon announces the availability of Amazon Nova 2 Sonic, our speech-to-speech model for natural, real-time conversational AI.  It offers best-in-class streaming speech understanding with robustness to background noise and users’ speaking styles, efficient dialog handling, and speech generation with expressive voices that can speak natively in multiple languages (Polyglot voices). It has superior reasoning, instruction following, and tool invocation accuracy over the previous model. Nova 2 Sonic builds on the capabilities introduced in the original Nova Sonic model with new features including expanded language support (Portuguese and Hindi), polyglot voices that enable the model to speak different languages with native expressivity using the same voice, and turn-taking controllability to allow developers to set low, medium, or high pause sensitivity. The model also adds cross-modal interaction, allowing users to seamlessly switch between voice and text in the s