OpenAI Launches ‘IndQA’ Benchmark

On 4th November 2025, OpenAI announced the launch of IndQA, a new benchmark designed to evaluate AI systems on their understanding of Indian languages, cultural nuances, and communication patterns.

Key Points

Collaborative Development: IndQA was developed with inputs from over 250 Indian experts across languages and domains, comprising 2,278 questions covering 12 languages and 10 cultural disciplines.
Covered Languages: The benchmark includes Bengali, English, Hindi, Kannada, Marathi, Odia, Telugu, Gujarati, Malayalam, Punjabi, Tamil, and Hinglish.
Cultural Domains: IndQA focuses on topics such as Architecture & Design, Arts & Culture, Everyday Life, Food & Cuisine, History, Law & Ethics, Literature & Linguistics, Media & Entertainment, Religion & Spirituality, and Sports & Recreation.
India as a Starting Point: OpenAI stated that India was a natural choice for launching this initiative, given its diverse linguistic landscape and the fact that ChatGPT’s second-largest user base comes from India.
Performance Evaluation: IndQA is used to track the performance of OpenAI’s frontier models over time. The company reported significant improvements in Indian language performance, while acknowledging scope for further enhancement.