AWS Announces Major Updates to Amazon Bedrock, AgentCore, and Strands AI Platforms
By Mr.Xu
Published:
Summary:In September 2026, AWS announced significant updates to its AI development tool platforms, including Amazon Bedrock, AgentCore, and Strands. These updates expand support for leading-edge models in Amazon Bedrock, optimize the runtime efficiency and reduce latency for serverless agents in AgentCore, and introduce a more efficient agent toolkit in Strands. The enhancements aim to provide developers with more flexible, efficient, and cost-effective AI solutions while strengthening the security and
AWS Announces Major Updates to Amazon Bedrock, AgentCore, and Strands AI Platforms
In September 2026, AWS unveiled comprehensive upgrades to its AI development tool platforms, focusing on three core products:
1. Amazon Bedrock
- Expanded Model Selection: Added support for OpenAI's Astra, Sol, and Luna models, offering developers a broader range of choices to meet diverse workload requirements.
- GPT-6 Astra: The flagship model, focusing on complex decision-making, document analysis, and software development, supports up to 1 million input tokens.
- GPT-6 Astra Ultrafast: Provides 6x faster inference for applications where speed is critical.
- Sol and Luna: Suitable for coding, computing, and professional workloads, balancing intelligence and cost.
- Enterprise-Grade Features: Supports more efficient knowledge base synchronization and automated data source connectors (e.g., ServiceNow, Confluence, Salesforce, Zendesk) to help enterprise AI agents retrieve the latest information.
2. Amazon Bedrock AgentCore
- Serverless Agent Optimization: Improved memory management and reduced cold start latency enable serverless agents to run faster and at lower cost.
- Efficient Development: Provides built-in human-in-the-loop workflows and durable sessions to reduce development overhead and mitigate risks.
3. Strands
- Open Source Agent Toolkit: Strands Harness is a new open source agent toolkit that reduces token usage by 28% while maintaining high accuracy.
- Local Decision Model: Strands Decider 2B is a lightweight 2B parameter decision model that makes quick decisions locally, suitable for tasks like tool selection, routing, and guardrails.
Technical Highlights
- Performance and Cost Optimization: AWS has optimized AI agent performance and cost-efficiency by improving memory management and reducing latency.
- Enterprise-Grade Security and Scalability: Amazon Bedrock supports enterprise-grade data source connectors and provides IAM permission management and auditing capabilities to ensure data security and compliance.
- Open Source and Flexibility: Strands Harness offers developers a flexible open source toolkit that can be customized and extended as needed.
Industry Impact and Developer Recommendations
- Accelerated AI Application Development: These updates enable developers to build and deploy AI applications more quickly while reducing development and operational costs.
- Enterprise AI Transformation: Enterprises can leverage these tools to achieve smarter automated decision-making, improving operational efficiency and customer experience.
- Developer Recommendations: Developers are advised to explore the latest features of Amazon Bedrock and Strands and try out the new open source toolkit to take full advantage of these platforms.
Conclusion
With these updates, AWS has further solidified its leadership in the AI development tool space, providing enterprises and developers with more powerful AI solutions. These tools not only enhance the performance and efficiency of AI agents but also strengthen the security and scalability of enterprise-grade applications, laying a solid foundation for an AI-driven future.
— END —Source: AWS Machine Learning Blog (2026-10-09)
Community Comments