OpenAI adds India-based processing for eligible API customers
Eligible organisations can run supported GPT-6 Luna API requests in India, extending residency from storage to model processing.
OpenAI has introduced India-based processing for eligible customers using its direct application programming interface. The option lets organisations run supported GPT-6 Luna requests in India, where the model computation for those requests takes place.
The change builds on OpenAI's existing regional storage option. Storage residency determines where eligible content is stored; inference residency determines where the model processes supported requests. The two are configured at the API project level.
OpenAI's developer documentation says the India processing option requires an organisation to have active Zero Data Retention. Supported requests can use the Responses and Chat Completions APIs. Hosted tools are outside the India regional processing offering described in those docs.
