OpenAI Launches Local Enterprise AI Inference in India

OpenAI has introduced local processing and inference for its application programming interface within India, according to The Times of India. The technical rollout permits qualifying commercial enterprises and software engineers to run supported artificial intelligence workloads directly on domestic server hardware. Under the new operating setup, incoming developer calls and the algorithmic steps creating answers occur entirely inside national boundaries. Client data and related materials will similarly remain housed domestically, enabling corporations to meet rigorous regulatory mandates. The development establishes domestic model computation as a core operational capability for organizations building machine learning applications in India.

Supported Workloads and API Integration Workflow

The domestic compute service debuts with support for the GPT-6 Luna model, catering to entities previously hindered by cross-border data transfer limitations. This arrangement directly accommodates financial services corporations facing explicit statutory localisation rules, as outlined by OpenAI. Direct API consumers can utilize domestic execution for GPT-6 Luna prompts and outputs across the Responses and Chat Completions interfaces. Teams activate local computation by setting up a fresh project workspace, choosing India for storage and compute locations, and designating the territorial gateway. The sovereign infrastructure aids standard document workflows such as condensing detailed records, preparing correspondence, and assisting workers with quantitative synthesis. OpenAI's API pricing page lists GPT-6 Luna at $0.10 for input and $0.50 for output.

Executive Perspectives on Enterprise AI Adoption

Certain domestic enterprises had postponed integrating AI into important operational tasks because protocols required local data handling, explained Nitin Bawankule, Head of Enterprise for India at OpenAI. Territorial execution provides organizations improved oversight regarding operational environments and lowers practical entry hurdles, Bawankule noted. He emphasized that localized infrastructure provides builders and emerging startups wider scope to design tools for internal and commercial deployments. The facility extends residency guarantees beyond where records reside, ensuring that model algorithms generating answers run on servers operating inside the country.

OpenAI for India Initiative and Infrastructure Alliances

The update broadens a data residency program started in May 2025 that originally permitted Indian clients to store files within national borders. The regional infrastructure push coincides with OpenAI's expanding footprint across the Indian technology market. Earlier this year, the provider introduced OpenAI for India, an ecosystem program focusing on compute hardware, corporate adoption, skills education, and commercial ties. As part of that broader program, OpenAI established a technical alliance with Tata Group to construct domestic server capacity. OpenAI's API inference residency is also available in other regions, with India, Australia and Japan the most recent additions, and AWS added regional inference for OpenAI models on Bedrock in August 2026.