|
Today, at the AWS Summit in New York, Sivasubramanian, VP agents AI AI, a daily main comment on how webling customers deform AI agents ready to make a scale. Here is our rounding of the biggest announcement of the event:
Introducing Amazon Bedrock Agentcore: Safely deploy and run AI agents on any scale (preview)
Amazon Bedrock Agentcore enables fast deployment and scaling of AI agents with business security. It provides memory management, identity control and the development of tool integration when working with any open source framework and endowment model.
Annonncing Amazon Nova Adaptation in Amazon Sagemaker AI
The AWS now allows extensive adaptation of Amazon Nova Foundation models via Sagemaker AI at all stages of model training. These skills, which are available as recipes, use SageMaker recipes, allow customers to adapt models to nova understanding and post-training, include recipes for fine-tuning and alignment with business security requirements in industries.
AWS FREE: New customers can start and explore AWS with credits up to $ 200
AWS improves its free level with $ 200 for credits for new users: $ 100 for registration and another $ 100 earned by completing activities with services such as Amazon EC2, Amazon Bedrock and AWS budgets.
Tweelvelabs video understanding models are now available in Amazon Bedrock
Twelvelabs video understanding models are now available on Amazon Bedrock and allow customers to search videos, classify scenes, summarize content and extract knowledge with accuracy and resurrection.
Amazon S3 metadata now metadata for all your S3 objects
Amazon S3 metadata now provides visibility to all S3 objects through live tables inventory and diary, allowing an analysis based on SQL existing and new objects with automatic updates within an hour of changes.
Introduction Amazon S3 Vectors: The first cloud storage with native vector support in scale (preview)
Amazon S3 Vectors is a new cloud object store that provides native support for storing and interviewing vectors on a massive scale and offers up to 90% reduction in costs compared to conventional approaches, while it integrates with Amazon Bedrock Nowlege, Sagemaker and OPSearch for AI.
Simplify the way from data to knowledge with new Amazon Sagemaker capacity
Amazon Sagemaker introduced three new capacities – Amazon QuickSight Integration for Dashboard, Management and Sharing, Amazon S3 Non -Transitive Data for Cataloging and Media Files, and Automatic deck data from Lake – Tathat Data Silos – Tathat Data Silos – Tathat Data Sow – tathat data silos – tathat data silos – tathat data silos – visualization and governance in individual experience.
Monitor and tune in applications controlled with new Amazon Eventbridge logging
Amazon Eventbridge now offers an improved logging ability that provides tracking of events, helps users to monitor and violate their applications based on detailed protocols that appear in publication, corresponding to the rules, delivered by subcuber failure.
Amazon EKS allows ultra -scales of AI/ml workload with support for 100k nodes per cluster
Amazon EKS now changes to 100,000 nodes per cluster, allowing massive workload AI/ml with up to 1.6 m AWS Accelerators or 800K NVIDIA GPUs. This allows organizations to effectively train and operate large AI models while holding Kubernetes compatibility and existing tool integration.