{
  "id": 3791782,
  "title": "Introducing India cross-Region inference for OpenAI GPT-5.6 models on Amazon Bedrock",
  "url": "https://urgent.news/2026/08/27/introducing-india-cross-region-inference-for-openai-gpt-5-6-models-on",
  "topic": "ai",
  "section": "AI",
  "published": "2026-08-27T18:36:08.000Z",
  "source": {
    "name": "AWS Machine Learning",
    "slug": "aws-machine-learning",
    "url": "https://aws.amazon.com/blogs/machine-learning/introducing-india-cross-region-inference-for-openai-gpt-5-6-models-on-amazon-bedrock/"
  },
  "original_language": "en",
  "account": "Amazon Bedrock now offers the OpenAI GPT-5.6 models, Terra and Luna, with India geographic cross-Region inference. Companies in India, particularly those in financial services, healthcare, and the public sector, can now utilize these models at scale while ensuring data stays within the country. Both models provide a 1-million-token context window, accept both text and image input, and generate text output.\n\nIndia geographic cross-Region inference allows inference requests to automatically route through multiple AWS regions, improving throughput without the need for manual capacity management in each region. Within the India geography, requests are routed between the Asia Pacific (Mumbai) Region (ap-south-1) and the Asia Pacific (Hyderabad) Region (ap-south-2). This enables scalability to meet demand while adhering to data residency requirements.\n\nCross-Region inference is managed through inference profiles, which are called as model IDs and specify the model and the AWS regions Amazon Bedrock can route the request to. There are two India geographic inference profiles: in.openai.gpt-5.6-terra for GPT-5.6 Terra, and in.openai.gpt-5.6-luna for GPT-5.6 Luna. Amazon CloudWatch and AWS CloudTrail record log entries in the source region, ensuring monitoring stays in one place.\n\nFor new applications, the bedrock-runtime endpoint is recommended, as it supports the Bedrock-native InvokeModel and Converse APIs, the OpenAI-compatible Responses and Chat Completions APIs, and cross-Region inference. By default, data is encrypted and Amazon Bedrock does not store model inputs or outputs, except for certain models where content flagged by the abuse-detection classifiers is retained for offline abuse detection.\n\nGlobal cross-Region inference is also available, routing to supported commercial AWS regions worldwide for maximum capacity. However, for workloads with local data processing requirements, the India prefixed profiles should be used. To learn more about global cross-Region inference, refer to the Introduce cross-Region inference for OpenAI GPT-5.6 models on Amazon Bedrock guide.",
  "summary": "Amazon Bedrock now supports the OpenAI GPT-5.6 models, Terra and Luna, in India with India geographic cross-Region inference. If you have local data processing requirements, you can now use these models at scale while Amazon Bedrock keeps inference requests and data within India.",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 2,
    "also_reported_by": [
      {
        "outlet": "AWS Machine Learning",
        "title": "Reduce ASR inference costs by 75% with NVIDIA MPS on Amazon EC2",
        "url": "https://urgent.news/2026/08/27/reduce-asr-inference-costs-by-75-with-nvidia-mps-on-amazon-ec2",
        "published": "2026-08-27T16:05:10.000Z"
      }
    ]
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}