All news with #aws tag

🚀 Amazon SageMaker AI now supports serverless customization for the Qwen3.6 27B parameter model using supervised fine-tuning (SFT) and reinforcement fine-tuning (RFT). This extends SageMaker's existing fine-tuning support for Qwen3.5 and other open-weight models. Serverless customization removes infrastructure management—SageMaker handles provisioning and orchestration—so teams pay only for what they use. The feature is available in US East (N. Virginia), US West (Oregon), Asia Pacific (Tokyo), and EU (Ireland).

🛠️ AWS announced general availability of the agent builder toolkit Kiro Power for AWS Transform, enabling partners and customers to create customized transformation agents tailored to modernization projects. The toolkit supports the full agent lifecycle—build with Kiro Power, share across teams and partner networks, and register agents with AWS Transform for discovery. It's available in the Kiro Power marketplace and targets Migration and Modernization Competency Partners, ISVs, and customers.

🛠️ AWS has made its AWS Transform agents available through an agent plugin, a Kiro Power in the Kiro marketplace, and the AWS Transform MCP server. Developers can now invoke transformation capabilities from their IDE, the web console, or programmatically, maintaining consistent job state across surfaces. IAM role authentication is supported, enabling use of existing AWS credentials for environments, workspaces, and jobs.

🔁 Amazon Aurora DSQL now supports change data capture (CDC) in preview, enabling real-time streaming of database changes directly to Amazon Kinesis Data Streams. The fully managed capability captures insert, update, and delete operations as discrete change events and removes the need to build and maintain custom streaming pipelines. It is designed for zero impact on database throughput and latency, requires no infrastructure setup, and integrates with AWS Lambda and Data Firehose for downstream delivery. CDC streaming in preview is available in all Regions where Aurora DSQL is offered and is billed using Distributed Processing Units (DPUs); standard Amazon Kinesis pricing applies separately.

🔔 AWS has added GLM-5.1-FP8 (from Z.ai) and Phi-4-mini-instruct (from Microsoft) to Amazon SageMaker JumpStart, expanding foundation model choices for enterprise workloads. GLM-5.1-FP8 targets agentic software engineering and multi-round optimization for repository-level code, debugging, and long-horizon automation. Phi-4-mini-instruct provides compact, low-latency reasoning across 24 languages and supports function calling for edge and latency-sensitive use cases. Customers can deploy these models via SageMaker Studio or the SageMaker Python SDK in a few clicks.

🔊 AWS has added three Qwen speech foundation models—Qwen3-TTS-12Hz-1.7B-CustomVoice, Qwen3-TTS-12Hz-1.7B-Base, and Qwen3-ASR-1.7B—to Amazon SageMaker JumpStart. The models deliver multilingual text-to-speech and automatic speech recognition capabilities across more than 10 languages and 52 languages/dialects. CustomVoice offers instruction-driven control over timbre, emotion, and prosody while Base enables 3-second rapid voice cloning. SageMaker JumpStart lets customers deploy these models from SageMaker Studio or via the SageMaker Python SDK with a few clicks.

🆕 AWS added FLUX.2-klein-base-4B and Qwen3-Embedding-0.6B to Amazon SageMaker JumpStart. FLUX.2 targets real-time image generation and multi-reference editing in a compact architecture that can run on consumer GPUs with about 13GB VRAM. Qwen3-Embedding delivers instruction-aware, multilingual text embeddings across 100+ languages for retrieval, RAG, and semantic search. Customers can deploy these models via SageMaker Studio or the SageMaker Python SDK.

🛡️ Amazon Connect Customer now offers a permission that lets agents access only their own performance evaluations directly in the Connect UI. Agents can search for contacts where they received evaluations, view evaluations alongside call recordings and transcripts, and submit an acknowledgment after review. Administrators can grant department-level contact visibility for investigations while preventing access to peers' evaluation data. The feature is available in all AWS regions where Amazon Connect Customer is offered.

🖥️ Amazon EC2 X8aedz instances are now available in Europe (Ireland), powered by 5th Gen AMD EPYC processors delivering up to 5 GHz single-thread performance. Built on sixth generation AWS Nitro Cards, these instances are optimized for EDA workloads and relational databases that need high CPU frequency and large memory. They provide a 32:1 memory-to-vCPU ratio, sizes from 2–96 vCPUs with up to 3,072 GiB memory and up to 8 TB local NVMe, and are offered via Savings Plans, On-Demand, and Spot.

🔒 The PQC Readiness Scanner automates inventory and continuous monitoring of AWS-terminated TLS endpoints — Application Load Balancer (ALB), Network Load Balancer (NLB), and Amazon API Gateway — to evaluate TLS policies for Post-Quantum Cryptography (PQC) readiness. It classifies endpoints into a three-tier framework (Tier 1: PQ-ready, Tier 2: PQ-ready with backward compatibility, Tier 3: not PQ-ready) and returns COMPLIANT/NON_COMPLIANT results with policy recommendations. Built as an AWS Config conformance pack with custom rules and Lambda functions, it supports organization-wide deployment via CloudFormation StackSets and S3-hosted artifacts. The scanner reduces manual review, tracks migration progress across accounts, and helps prioritize upgrades to TLS 1.3 with PQC key exchange.

🛠️ The AWS Transform agentic AI assistant is integrated into the AWS Toolkit for Visual Studio, giving .NET developers an in‑IDE, conversational workflow to assess, plan, and execute application modernization. It provides visibility, checkpointing, interactive diffs, automatic build-fix attempts, detailed worklogs, and downloadable HTML reports while preserving context between the web console and Visual Studio. Agents are also accessible via Kiro and other AI coding environments and are available in multiple AWS Regions.

🔗AWS announced that RTB Fabric now supports custom domains for real-time bidding transactions received through external links. This enables AdTech companies to preserve existing public endpoints (for example, bid.company.com/path) by using their own DNS and CNAME records without forcing partners to reconfigure endpoints. Customers can create routing rules to map URL patterns to specific RTB Fabric links so demand- and supply-side platforms can route partner traffic seamlessly. The capability is available in all AWS Regions that support RTB Fabric and maintains low latency while reducing standard cloud networking costs.

🔒 The updated AWS User Guide to Governance, Risk, and Compliance for Responsible AI Adoption provides Financial Services customers practical GRC guidance for deploying AI responsibly. It covers governance, risk management, compliance, data and model management, and AI agent oversight, and maps these considerations to AWS capabilities. The guide highlights services such as Amazon Bedrock AgentCore, Bedrock Guardrails, Bedrock Agents, SageMaker Autopilot, and SageMaker Model Monitor. It complements existing AWS responsible AI and Well-Architected resources and is available on the AWS Whitepaper portal.

🗄️Amazon announced that Amazon FSx for OpenZFS now supports creating Multi-AZ file systems from participant accounts in shared VPCs, enabling organizations to decentralize storage administration while keeping network control centralized. Previously, participant accounts could only create Single-AZ file systems in shared VPCs and needed to own the VPC for Multi-AZ deployments. This change allows participant accounts to create any FSx for OpenZFS file system in a shared VPC across all Regions where the service is available, improving high availability and operational flexibility.

🔒 AWS announced the completion of PCI PIN and PCI P2PE assessments for AWS Payment Cryptography, expanding validations to include Key Management (KMCP) and Key Loading (KLCP) alongside the existing Decryption Management (DMCP). The coverage is extended to South America (São Paulo) and Asia Pacific (Sydney) Regions. These attestations allow customers to use PCI PTS HSM-certified, AWS-managed HSMs with compliant key management to simplify regulated deployments.

🔍 AWS Security Agent now offers a preview of full repository code review, an AI-driven capability that performs deep, context-aware analysis across entire repositories. It models application architecture, trust boundaries, and data flows rather than relying on pattern matching, and returns developer-ready findings with structured evidence and concrete remediation. The feature is designed to complement existing SAST tools and is available in preview at no additional charge while AWS solicits customer feedback.

🚀 Amazon EventBridge Scheduler expands its AWS SDK integrations with 619 new API actions across 13 additional services, including support for AWS Lambda Managed Instances. This update enables customers to schedule direct API invocations for a broader set of AWS services without writing custom integration code. As a serverless scheduler, EventBridge Scheduler can manage billions of scheduled events and now supports time-based scaling of Lambda managed instances for more precise capacity control. The enhancements are generally available in all Regions where Scheduler is offered, subject to target service availability.

🔒 AWS today introduced full repository code review in AWS Security Agent, a capability that performs deep, context-aware security analysis across entire codebases. Unlike traditional static scanners, it reasons about architecture, trust boundaries, and data flows to surface systemic vulnerabilities. When issues are identified, the scanner generates file- and line-specific remediation guidance and exploit proofs-of-concept to accelerate fixes; preview access is available at no extra charge in all Regions.

🔒 Amazon SageMaker Feature Store now supports the SageMaker Python SDK v3, providing modular APIs to manage feature groups with less boilerplate. Data scientists can enable Lake Formation access controls to enforce column- and row-level permissions on offline store data at feature group creation. The SDK also exposes Apache Iceberg table properties for configuring compaction and snapshot expiration to optimize storage and queries. Available in all AWS Regions where Feature Store is offered; install v3.8.0 or later to begin.

🔁 Amazon EKS now supports Amazon Application Recovery Controller (ARC) zonal shift and zonal autoshift when using the open-source Karpenter for compute provisioning. ARC automates redirecting in-cluster network traffic away from an impaired AZ and can perform practice runs to validate cluster behavior. During a zonal shift, Karpenter stops provisioning in the impacted AZ, halts voluntary disruptions there, and avoids scheduling actions that depend on that zone. Enable support by setting ENABLE_ZONAL_SHIFT.