Run Generative AI inference with Amazon Bedrock in Asia Pacific (New Zealand)

2026-03-26 · Source: Artificial Intelligence · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Cloud Computing & IT Infrastructure · Depth: Advanced, medium

Summary

Amazon Bedrock is now available in the Asia Pacific (New Zealand) Region (ap-southeast-6), enabling customers to access foundation models (FMs) like Anthropic Claude (Opus 4.5, Opus 4.6, Sonnet 4.5, Sonnet 4.6, Haiku 4.5) and Amazon Nova 2 Lite models directly from Auckland. This launch introduces cross-Region inference, a capability that distributes inference processing across multiple AWS Regions to enhance throughput. Requests from Auckland can utilize either AU geographic cross-Region inference, routing within Auckland, Sydney (ap-southeast-2), and Melbourne (ap-southeast-4) for data residency, or global cross-Region inference, routing to supported commercial AWS Regions worldwide for higher throughput and resilience. All data remains on the AWS network, is encrypted in transit, and requests are logged in AWS CloudTrail in the source Region.

Key takeaway

For AI Architects and MLOps Engineers deploying foundation models in New Zealand, this Amazon Bedrock expansion allows local API calls while leveraging cross-Region inference for performance or data residency. You should evaluate whether AU geographic routing meets your data residency needs for Anthropic Claude models or if global routing is necessary for broader model access and higher throughput. Configure IAM and SCPs carefully to align with your security and compliance requirements.

Key insights

Amazon Bedrock's cross-Region inference in New Zealand enhances FM access, throughput, and data residency options.

Principles

• Data residency can be maintained via geographic routing.
• Global routing offers higher throughput and resilience.
• Least privilege applies to IAM policies for cross-Region access.

Method

Invoke FMs via cross-Region inference profiles (e.g., "au." or "global." prefixes) from a source Region, with requests routed to destination Regions for processing, ensuring data encryption and logging.

In practice

• Use "au." prefix for ANZ data residency.
• Use "global." prefix for maximum throughput.
• Configure IAM policies for source and destination Regions.

Topics

• Amazon Bedrock
• Cross-Region Inference
• Asia Pacific (New Zealand) Region
• Anthropic Claude
• AWS IAM

Best for: AI Engineer, MLOps Engineer, AI Architect

Related on AIssential

• Access Anthropic Claude models in India on Amazon Bedrock with Global cross-Region inference — Amazon Bedrock's Global CRIS enhances generative AI inference by distributing Anthropic Claude models across AWS Regions for improved scale and resilience.
• Understanding Amazon Bedrock model lifecycle — Proactive management of Amazon Bedrock's FM lifecycle is crucial for maintaining AI application continuity and leveraging updated capabilities.
• OpenAI models now available on Amazon Web Services — OpenAI's GPT-5.5, GPT-5.4, and Codex are now available on AWS Bedrock, integrating with existing AWS infrastructure and security.
• AWS introduces Claude Platform to global customers — AWS now offers direct access to Anthropic's native Claude Platform, streamlining integration and feature access.
• AWS adds GPT-5.5, GPT-5.4 and Codex to Amazon Bedrock — AWS Bedrock now offers OpenAI's GPT-5.5, GPT-5.4, and Codex, providing enterprise access with integrated billing and unified APIs.
• Claude Fable 5 on Bedrock Requires Sharing Inference Data with Anthropic — The introduction of Claude Fable 5 on Bedrock fundamentally altered data residency guarantees, requiring inference data sharing with Anthropic for safety review.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Artificial Intelligence.