OpenAI launched API inference residency in India on Thursday, letting eligible business customers run model computation for GPT-6 Luna inside the country instead of sending requests overseas. The feature covers requests made through the Responses and Chat Completions APIs, and supported customer content is also stored within India.
The launch extends OpenAI's existing data residency offering in India, which until now covered storage only. With inference residency, API requests are processed and the model computation itself runs in India. That matters for businesses bound by data localisation rules, with financial services firms among the customers the change is aimed at.
GPT-6 Luna is the only model supported at launch. OpenAI describes Luna as its most efficient model for focused, high-volume tasks, and the in-country use cases cited include document summarisation, drafting responses and analysing information.
Key facts
- 01OpenAI launched API inference residency in India on November 12, 2026, starting with GPT-6 Luna.
- 02Requests through the Responses and Chat Completions APIs are processed in India, with supported customer content stored there.
- 03Access requires active Zero Data Retention and standard pay-as-you-go processing; hosted tools are excluded.
- 04Data residency endpoints carry a 10% price uplift for eligible models released on or after March 5, 2026.
- 05India joins Australia and Japan as regions where OpenAI offers in-country processing, not just storage.
Setup runs through a new API project. Customers select India for both data storage and processing, then send requests to the India regional endpoint at in.api.openai.com, according to OpenAI's platform documentation.
The option comes with conditions. Regional processing in India requires an active Zero Data Retention arrangement, and modified abuse monitoring alone does not qualify. Only standard pay-as-you-go processing is supported, hosted tools are not available for in-region processing, and access is arranged through OpenAI's sales team under a modified retention amendment.
Pricing carries a premium. OpenAI applies a 10% price uplift on data residency endpoints for eligible models released on or after March 5, 2026. Residency also applies to customer content rather than system data, which may still be processed or stored outside the selected region.
Nitin Bawankule, OpenAI's head of enterprise for India, said some Indian businesses have not fully deployed AI for important work because they need the information involved to be processed within the country.
“Local inference gives them more control over where supported workloads run and helps address a practical barrier to adoption.”— Nitin Bawankule, Head of Enterprise - India, OpenAI
India now sits alongside Australia and Japan as a region where OpenAI offers in-country processing, not just storage. Indian customers already had one route to local inference for OpenAI models: AWS added an India geographic inference profile on Amazon Bedrock in August 2026 for GPT-5.6 Terra and Luna.
The rollout is part of OpenAI's broader OpenAI for India effort, which includes a partnership with Tata Group to build domestic AI data centre capacity.
At launch, inference residency in India is limited to GPT-6 Luna workloads called directly through the API.
Working on something we should cover, or seeing a story we missed? Send leads, documents, or feedback to hello@aichatdaily.com. For sensitive tips, see our secure tips page for Signal and PGP options.
Spotted an error? Email hello@aichatdaily.com with the URL and the issue, or read our full corrections policy.
