Introducing OpenAI models on Amazon Bedrock for in-country inferencing in India

Amazon Bedrock has added support for OpenAI’s GPT‑5.6 family—specifically the Terra and Luna models—within India, enabling cross‑Region inference that keeps all request traffic and data confined to Indian data centers. This move allows customers to run large‑scale LLM workloads locally without routing traffic outside the country, a key requirement for many compliance‑heavy industries. While the announcement does not disclose latency or cost figures, the geographic isolation may introduce higher round‑trip times compared to global endpoints, yet it satisfies strict data residency mandates.

⚡ Key Takeaways

• OpenAI GPT‑5.6 models Terra and Luna are now available on Amazon Bedrock in India.
• Bedrock’s India‑centric cross‑Region inference architecture guarantees that inference requests and data never leave Indian borders.
• The new offering may incur higher latency relative to global Bedrock endpoints, but it fulfills local data‑processing compliance.
• To use the models, specify the `ap-south-1` region in the Bedrock client and invoke the GPT‑5.6 inference API.
• Availability is limited to the Indian region; requests from other regions will not be routed to these models.
• WhyItMatters: Engineers building AI services that must comply with India’s data‑residency laws can now deploy state‑of‑the‑art GPT‑5.6 models without compromising on scale or performance.
• TechnicalLevel: Intermediate
• TargetAudience: ML Engineers
• PracticalSteps:
• Configure the Bedrock client with `region='ap-south-1'`.
• Call the `invokeModel` API endpoint for the GPT‑5.6 (Terra or Luna) model.
• Verify that your data pipeline routes all payloads

Engineers building AI services that must comply with India’s data‑residency laws can now deploy state‑of‑the‑art GPT‑5.6 models without compromising on scale or performance.

✅ Practical Steps

1. Configure the Bedrock client with `region='ap-south-1'`.
2. Call the `invokeModel` API endpoint for the GPT‑5.6 (Terra or Luna) model.
3. Verify that your data pipeline routes all payloads

Want the full story? Read the original article.