# AWS Launches GPT-5.6 Access in Australia Three distinct model profiles now available via Amazon Bedrock. - **Source:** [Unite.AI](https://www.unite.ai) - **Published:** 2026-09-02 - **Topic:** [Data & AI](https://brifing.si/c/data-ai) - **Original article:** - **On Brifing:** https://brifing.si/s/1792400a-4ac4-4f77-a9aa-c4bea9f2ae25 ## GPT-5.6 Access in Australia AWS announced on September 2, 2026, that developers in Australia can now access OpenAI's GPT-5.6 models through Amazon Bedrock. Requests from Sydney (ap-southeast-2) and Melbourne (ap-southeast-4) regions are routed to supported commercial AWS regions via global cross-region inference. This setup provides access to a broader capacity pool without requiring applications to manage destination region routing. ## Model Variants and Features Three GPT-5.6 variants are offered: Sol, optimized for demanding reasoning, coding, and agentic workloads; Terra, balancing performance and cost for everyday production use; and Luna, providing fast, affordable inference for high-volume, latency-sensitive applications. All models accept text and image inputs, generate text, and support context windows of up to 1 million tokens. ## Access Paths and APIs Developers can invoke the models through three access paths on the Bedrock Runtime endpoint: OpenAI Responses API, OpenAI Chat Completions API, and Amazon Bedrock Converse API. The OpenAI-compatible APIs are called on the endpoint's /openai/v1 paths instead of through AWS SDKs. The endpoint accepts either AWS Signature Version 4 signing or an Amazon Bedrock model inference API key. ## Prompt Caching Options GPT-5.6 supports prompt caching in two modes through its APIs. Implicit caching is enabled by default with no code changes, while explicit caching allows developers to define reusable prefixes, cache boundaries, and cache keys. AWS cautions that profile membership and model availability are subject to change. ## Authentication and Security For organizations federating identity through Okta, Auth0, Microsoft Entra ID, Amazon Cognito, or AWS IAM Identity Center, AWS provides a sample credential helper that exchanges OpenID Connect tokens for temporary AWS credentials. Integrated into the latest Codex CLI, this feature enables secure inference without API keys. ## Deployment Prerequisites Australian deployments require an AWS account with Sydney or Melbourne enabled as the source region, an IAM role/user with permissions to invoke GPT-5.6 inference profiles, and Python 3.9+ with openai, boto3, and aws-bedrock-token-generator packages. Organizations using service control policies must verify their policy allows GPT-5.6 global inference profiles in the selected source region. ## Quota Management and Performance GPT-5.6 on-demand quotas are measured in requests per minute and tokens per minute. Input tokens and cache-write input tokens count at a 1:1 rate, while each output token consumes 10 tokens from the quota. Quotas can be reviewed and increased via the Service Quotas console in the application's source region. ## Monitoring and Analytics As GPT-5.6 requests use the Bedrock Runtime API, calls through global inference profiles appear in model invocation logs. Codex exports metrics via OpenTelemetry protocol, and CloudWatch Coding Agent Insights provides a dashboard covering token usage, API requests, active users, conversation activity, and cache hit rate. ## Sources covering this story - https://www.unite.ai/aws-opens-gpt-5-6-access-on-amazon-bedrock-from-australian-regions/ — --- _Summary written by Brifing from the original reporting. Copyright in the original article belongs to its publisher; read the full piece at the link above._