All news with #amazon sagemaker ai tag

🚀 Amazon SageMaker Unified Studio now supports AWS Glue 5.1 for Visual ETL, notebook, and code-based data processing jobs. With Glue 5.1 you can run on Apache Spark 3.5.6 with Python 3.11 and Scala 2.12.18, and use updated open table formats including Apache Iceberg 1.10.0, Apache Hudi 1.0.2, and Delta Lake 3.3.2. Select Glue 5.1 from the job version dropdown to apply the runtime across Visual ETL, notebooks, and code jobs.

🔗 AWS now enables Kiro IDE to connect remotely to Amazon SageMaker Unified Studio, allowing data scientists, ML engineers, and developers to use their local Kiro setup — including spec-driven development, conversational coding, and automated feature generation — while running workloads on SageMaker’s scalable compute. The integration uses the AWS Toolkit extension for secure IAM-based authentication and preserves local specs, steering files, and hooks. This reduces context switching and keeps agentic development workflows within a single environment across AWS analytics and ML services. The capability is available in all Regions where SageMaker Unified Studio is offered.

🔧 Amazon SageMaker HyperPod now supports API-driven Slurm configuration, enabling you to define Slurm topology, instance group to partition mappings, and FSx filesystem mounts directly in the cluster CreateCluster and UpdateCluster APIs or via the AWS Console. The update lets you specify node roles such as Controller, Login, and Compute per instance group and mount FSx for Lustre or FSx for OpenZFS filesystems. A new SlurmConfigStrategy (Managed, Overwrite, Merge) detects partition-node drift and controls whether updates are paused, overwritten, or merged to preserve manual customizations.

🔧 Amazon SageMaker HyperPod now lets operators manage individual cluster nodes directly from the AWS Console. The console enables SSM session launches, copyable pre-populated SSM CLI commands, and direct node actions such as reboot, delete, and replace, with support for batch operations across multiple nodes. Available in all Regions where HyperPod is supported, these controls reduce context switching and speed manual recovery for time-sensitive AI training and inference workloads.

🔈 Cartesia's Sonic 3 model is now available in Amazon SageMaker JumpStart, giving AWS customers a turnkey option for advanced streaming text-to-speech. Sonic 3 is a state space model (SSM) offering high naturalness, accurate transcript following, sub-100ms latency, and fine-grained control over volume, speed, and emotion. It supports 42 languages, natural laughter, and voices optimized for agents and expressive characters. Deployments can be launched from SageMaker Studio or via the SageMaker Python SDK.

🔍 Amazon SageMaker now provides Data Lineage for Apache Spark jobs run on Amazon EMR and AWS Glue within IDC-based SageMaker Unified Studio domains. The feature captures schema and column-level transformations from EMR-EC2, EMR-Serverless, EMR-EKS, and Glue, and makes lineage explorable as a visual graph or queryable via APIs. Teams can compare transformation history across Spark jobs to investigate regressions, trace root causes, and assess impact. Spark lineage is available in all existing SageMaker Unified Studio regions.

📢 AWS has added DeepSeek OCR, MiniMax M2.1, and Qwen3-VL-8B-Instruct to SageMaker JumpStart, expanding the set of foundation models available to customers. DeepSeek OCR focuses on visual-text compression and structured extraction from forms, invoices, diagrams, and other dense document layouts. MiniMax M2.1 targets multilingual coding, tool use, instruction following, and long-horizon planning to support autonomous workflows. Qwen3-VL-8B-Instruct enhances vision-language reasoning, spatial and video dynamics comprehension, and extended context handling. Customers can deploy any of these models via the JumpStart catalog or the SageMaker Python SDK to accelerate AI application development on AWS infrastructure.

🚀 With Amazon SageMaker JumpStart, customers can now deploy four NVIDIA NIMs — ProteinMPNN, Nemotron-3.5B-Instruct, MSA Search NIM, and Cosmos Reason — with one click. These prebuilt, optimized inference microservices are designed for NVIDIA-accelerated infrastructure and target biosciences and physical AI use cases. They enable protein sequence optimization, GPU-accelerated multiple sequence alignment, large-context reasoning and agentic tool calling, and vision-language planning for robotics. Deployments are accessible from the SageMaker JumpStart catalog or via the SageMaker Python SDK.

🔍 Amazon SageMaker HyperPod now surfaces detailed error messages and points directly to the CloudWatch log group and log stream that captured lifecycle script output. You can view these messages through the DescribeCluster API or via the SageMaker console, which includes a 'View lifecycle script logs' button to open the exact CloudWatch stream. CloudWatch entries now contain execution markers (begin, download start/complete, success/failure) to help pinpoint where provisioning failed. This enhancement is available in all Regions where HyperPod is supported and reduces time to diagnose lifecycle script issues.

🧭 The Amazon SageMaker HyperPod console now validates AWS service quotas for your account before initiating cluster creation. The console automatically compares your requested cluster configuration—instance types, EBS volume sizes, and VPC-related resources—against account-level quotas and presents a clear table of expected utilization, applied quota values, and compliance status. If validation detects potential quota shortfalls, it issues a warning and provides direct links to the Service Quotas console so you can request increases before provisioning begins.

🚀 MiniMax-M2 is now available on SageMaker JumpStart, enabling immediate deployment of this efficient open-source MoE model in minutes. The model combines 230 billion total parameters with 10 billion active parameters to deliver a compact, fast, and cost-effective option optimized for coding and agentic tasks while preserving strong general intelligence. Customers can deploy via SageMaker Studio or the SageMaker Python SDK and follow AWS best practices for production use.

🚀Amazon announced that SageMaker AI is now available in the Asia Pacific (New Zealand) AWS Region. Starting today, developers and data scientists in New Zealand can build, train, and deploy machine learning models locally using the fully managed SageMaker AI platform. The service removes much of the operational overhead across the ML lifecycle, helping teams move from experimentation to production more quickly and consistently. Customers should review AWS documentation and pricing to get started.

🔁 Amazon SageMaker Notebook instances now support self-service migration via the PlatformIdentifier parameter in the UpdateNotebookInstance API. You can update unsupported platform identifiers (notebook-al1-v1, notebook-al2-v1, notebook-al2-v2) to supported versions (notebook-al2-v3, notebook-al2023-v1) while preserving data and configurations. The capability is available through AWS CLI (v2.31.27+) and SDKs in all Regions where Notebook instances are supported. This simplifies keeping instances current and reduces manual migration effort.

🚀 Amazon SageMaker HyperPod now supports checkpointless training, a foundational capability that eliminates the need for checkpoint-based, job-level restarts for distributed model training. Checkpointless training preserves forward training state across the cluster, automatically swaps out failed nodes, and uses peer-to-peer state transfer to resume progress, reducing recovery time from hours to minutes. The feature can deliver up to 95% training goodput at very large scale, is available in all Regions where HyperPod runs, and can be enabled with zero code changes for popular recipes or with minimal PyTorch modifications for custom models.

🚀 Amazon SageMaker AI now offers a serverless model customization capability that lets developers quickly fine-tune popular models using supervised learning, reinforcement learning, and direct preference optimization. The fully managed, end-to-end workflow simplifies data preparation, synthetic data generation, training, evaluation, and deployment through an easy-to-use interface. Supported base models include Amazon Nova, Llama, Qwen, DeepSeek, and GPT-OSS. The AI agent-guided workflow is in preview with regional availability and a waitlist.

⚡ Amazon SageMaker HyperPod now supports elastic training, automatically scaling distributed training jobs to absorb idle accelerators and contract when higher‑priority workloads require resources. This eliminates the manual cycle of halting jobs, reconfiguring parameters, and restarting distributed training, which previously demanded specialized engineering time. Organizations can start training with minimal resources and grow opportunistically, improving cluster utilization and reducing costs. Elastic training can be enabled with zero code changes for public models like Llama and GPT OSS, and requires only lightweight configuration updates for custom architectures.

🚀 Amazon Web Services announced general availability of Nova Forge, a SageMaker AI service that enables organizations to build custom frontier models from Nova checkpoints across pre-, mid-, and post-training phases. Developers can blend proprietary data with Amazon-curated datasets, run Reinforcement Fine Tuning (RFT) with in-environment reward functions, and apply custom safety guardrails via a built-in responsible AI toolkit. Nova Forge includes early access to Nova 2 Pro and Nova 2 Omni and is available today in US East (N. Virginia).

🚀 AWS AI Factories are now available to deploy high-performance AWS AI infrastructure inside customer data centers, combining AWS Trainium, NVIDIA GPUs, low-latency networking, and optimized storage. The service integrates Amazon Bedrock and Amazon SageMaker to provide immediate access to foundation models without separate provider contracts. AWS manages procurement, setup, and operations while customers supply space and power, enabling isolated, sovereign deployments that accelerate AI initiatives.

🔍 Amazon SageMaker Catalog now exports asset metadata as an Apache Iceberg table via Amazon S3 Tables, enabling teams to query catalog inventory with standard SQL without building custom ETL. The export includes technical fields (resource_id, resource_type), business metadata (asset_name, business_description), ownership details, and timestamps, partitioned by snapshot_date for time travel queries. The dataset appears in SageMaker Unified Studio and is queryable from Amazon Athena, Studio notebooks, AI agents, and BI tools. Available in all supported Regions at no additional SageMaker charge; you pay for S3 Tables storage and Athena queries.

🧠 Amazon SageMaker AI now offers a serverless MLflow capability that automatically scales to support experiment tracking and model development without infrastructure setup. The service scales up for demanding workloads and scales down during idle periods, reducing operational overhead. Administrators can enable cross-account access via Resource Access Manager (RAM). The feature integrates with SageMaker AI JumpStart, Model Registry, and Pipelines and is offered at no additional charge in select AWS Regions.