Exclusive Content:

Haiper steps out of stealth mode, secures $13.8 million seed funding for video-generative AI

Haiper Emerges from Stealth Mode with $13.8 Million Seed...

Running Your ML Notebook on Databricks: A Step-by-Step Guide

A Step-by-Step Guide to Hosting Machine Learning Notebooks in...

“Revealing Weak Infosec Practices that Open the Door for Cyber Criminals in Your Organization” • The Register

Warning: Stolen ChatGPT Credentials a Hot Commodity on the...

Run Generative AI Inference Using Amazon Bedrock in Asia Pacific (New Zealand)

Amazon Bedrock Now Available in New Zealand: A New Era for Cross-Region Inference

Unlocking Foundation Model Access for Kiwi Customers

Exciting News for New Zealand: Amazon Bedrock Introduces Cross-Region Inference!

Kia ora, New Zealand! We have thrilling updates for our customers in this beautiful country. After numerous requests, Amazon Bedrock is now available in the Asia Pacific (New Zealand) Region (ap-southeast-6). This means you can now access leading foundation models (FMs) from Anthropic and Amazon directly from Auckland with the added capability of cross-region inference!

In this post, we will delve into how cross-region inference operates in New Zealand, the models accessible via geographic and global routing, and guide you on how to make your first API call.

Key Topics Covered:

  1. Understanding Cross-Region Inference
  2. What’s New: New Zealand as a Source Region
  3. Getting Started
  4. Security and Compliance Considerations
  5. Auditing and Monitoring
  6. Conclusion

Understanding Cross-Region Inference

Cross-region inference is an innovative feature of Amazon Bedrock that distributes inference processing across multiple AWS Regions, enhancing throughput and scalability. When making a request, Amazon Bedrock routes it from your source Region (where the API call is made) to a destination Region (where processing occurs).

This functionality operates exclusively over the AWS network, ensuring data transmission is secure and encrypted. Each cross-region inference request is logged in AWS CloudTrail, maintaining a clear audit trail.

The Two Types of Cross-Region Inference:

  • Geographic Cross-Region Inference: It routes requests within a defined geographic boundary. For example, requests from Auckland will flow through Auckland, Sydney, and Melbourne.

  • Global Cross-Region Inference: It allows routing to supported AWS Regions worldwide, maximizing throughput for organizations that don’t have stringent data residency requirements.

What’s New: New Zealand as a Source Region

With Amazon Bedrock’s launch in New Zealand, Auckland now serves as a new source Region for both geographic and global cross-region inference. This means customers can initiate API calls locally while accessing a wide array of foundation models.

AU Geographic Cross-Region Inference Configuration

The following table outlines source and destination Region routing:

Source Region Destination Regions Description
Auckland (ap-southeast-6) ap-southeast-6, ap-southeast-2, ap-southeast-4 New – Requests can be routed to Sydney, Melbourne, or remain in Auckland.

This increased flexibility allows better capacity distribution among the three locations.

Getting Started

Supported Models and Inference Profile IDs

Cross-region inference from New Zealand includes foundation models from various providers, ensuring that you always have access to the latest technologies. Here’s a snapshot:

Cross-Region Inference Type Example Models
AU Geographic Anthropic Claude Opus 4.6, Claude Sonnet 4.6
Global All models available under AU Geographic

To utilize these models, simply replace the foundational model ID with the geographic (au.) or global (global.) prefix.

IAM Permissions Configuration

To enable foundation model invocation through AU geographic cross-region inference, your AWS Identity and Access Management (IAM) policy should include the following:

  1. Permissions to access the inference profile in the source Region.
  2. Permissions to access foundation models in all destination Regions defined in the AU inference profile.

A sample IAM policy is illustrated below:

{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Effect": "Allow",
      "Action": ["bedrock:InvokeModel"],
      "Resource": "arn:aws:bedrock:ap-southeast-6::inference-profile/au.anthropic.claude-sonnet-4-6"
    }
  ]
}

Security and Compliance Considerations

Security is paramount. All requests travel securely over the AWS Global Network with end-to-end encryption. When operating with Service Control Policies (SCPs), ensure that actions in the Auckland Region are properly regulated to maintain compliance.

Auditing and Monitoring

AWS CloudTrail automatically logs all cross-region inference activities in the source Region, allowing you to audit requests effectively. Metrics provided through Amazon CloudWatch enable you to monitor your operational health, including invocation counts, latency, and error rates.

Conclusion

We are thrilled to introduce cross-region inference support from New Zealand on Amazon Bedrock. With this update, New Zealand customers can now enjoy local API calls, leveraging advanced foundation models without compromising security or compliance.

Key Takeaways:

  • Local Source Region: New Zealand customers can now initiate Amazon Bedrock API calls directly from Auckland.
  • ANZ Data Residency: Geographic inference keeps data processing within ANZ.
  • Global Model Access: Benefit from the highest throughput with global inference capabilities.

Next Steps:

To get started, simply:

  1. Sign in to the Amazon Bedrock console in the Auckland Region.
  2. Configure your IAM permissions.
  3. Make your first API call using the au. model ID.

For more information, please explore our resources and guides for detailed instructions and support.


About the Authors

Zohreh Norouzi: Security Solutions Architect at AWS, focusing on secure generative AI solutions.
Melanie Li, PhD: Senior Generative AI Specialist focusing on customer solutions in AI.
Saurabh Trikande: Senior Product Manager passionate about democratizing AI.
James Zheng: Software Development Manager at AWS.
William Yap & Julia Bodia: Principal Product Managers for Amazon Bedrock.

Get ready to elevate your AI capabilities with Amazon Bedrock!

Latest

What You Need to Know Before Seeking Medical Advice from ChatGPT to Ensure Safe Responses

The Rise of AI in Health Consultations: ChatGPT as...

Sperm’s Journey in Space: Australian Research Explores Microgravity Effects

Navigating Reproduction in Space: Challenges for Sperm in Microgravity Navigating...

Deploying Voice Agents Using Pipecat and Amazon Bedrock AgentCore Runtime – Part 1

Leveraging AWS and Pipecat to Build Intelligent Voice Agents:...

OpenAI Introduces ChatGPT Library for Long-Term User Storage Solutions

OpenAI Introduces File Storage and Retrieval Feature for ChatGPT...

Don't miss

Haiper steps out of stealth mode, secures $13.8 million seed funding for video-generative AI

Haiper Emerges from Stealth Mode with $13.8 Million Seed...

Running Your ML Notebook on Databricks: A Step-by-Step Guide

A Step-by-Step Guide to Hosting Machine Learning Notebooks in...

VOXI UK Launches First AI Chatbot to Support Customers

VOXI Launches AI Chatbot to Revolutionize Customer Services in...

Investing in digital infrastructure key to realizing generative AI’s potential for driving economic growth | articles

Challenges Hindering the Widescale Deployment of Generative AI: Legal,...

Scaling Video Insights with Amazon Bedrock’s Multimodal Models

Unlocking Video Insights: Harnessing the Power of Amazon Bedrock for Advanced Understanding The Evolution of Video Analysis Three Approaches to Video Understanding Frame-Based Workflow: Precision at Scale Shot-Based...

Deploy SageMaker AI Inference Endpoints with Configured GPU Capacity Using Training...

A Comprehensive Guide to Optimizing Inference Workloads with Amazon SageMaker AI Training Plans Introduction to LLMs and GPU Capacity Challenges Leveraging Amazon SageMaker for Predictable Inference...

Transforming Security Alerts with Reco and Amazon Bedrock

Transforming Security Alerts with AI: A Deep Dive into Reco’s Implementation of Amazon Bedrock Co-written by Tal Shapira and Tamir Friedman from Reco In this comprehensive...