AWS has announced that teams in Australia can now access OpenAI GPT-5.6 Sol, Terra, and Luna models through Amazon Bedrock using global cross-Region inference from the Asia Pacific (Sydney) and Asia Pacific (Melbourne) Regions. The accompanying post explains how to invoke the models, apply prompt caching, configure Codex with OpenID Connect authentication, and track usage with Amazon CloudWatch.
Why it matters
Regional availability lets Australian teams run these OpenAI models within nearby AWS Regions, which is relevant for organizations that need to manage where inference is served.
Who should care
Developers and enterprise teams in Australia building on Amazon Bedrock, particularly those interested in prompt caching, Codex integration, and CloudWatch-based monitoring.