Mumbai: Neysa.ai has expanded its catalog of open-weight AI models on its Velocis cloud platform, adding OpenAI’s GPT-OSS models to strengthen India’s sovereign AI mission. The move aims to reduce reliance on foreign-hosted, closed systems that have long posed challenges around compliance, innovation, and cost predictability for Indian organizations.
Until now, AI adoption in India has largely depended on overseas platforms, raising concerns about data security, transparency, and long-term control. By integrating GPT-OSS into its Velocis ecosystem, Neysa joins other open-weight providers such as Llama, DeepSeek, Qwen, and Mistral to offer models that can be deployed and scaled entirely within India. These models are available as Managed Inference Endpoints, enabling enterprises and developers to move from concept to production API quickly while ensuring compliance with local regulations.
Sharad Sanghi, Co-founder and CEO of Neysa, noted that building AI infrastructure locally is critical for protecting data and supporting India’s digital economy. Chief Product Officer Karan Kirpalani emphasized the importance of transparency and predictable economics, highlighting that the platform allows users to avoid token-based billing models and retain greater control over their deployments.
The expanded catalog includes models such as GPT-OSS-120B, designed for advanced reasoning, and GPT-OSS-20B, optimized for high-performance workloads, alongside fine-tuned variants like GPTQ and AutoRound. With this step, Neysa positions itself within the broader national effort to establish an indigenous AI ecosystem capable of managing the full lifecycle of AI applications securely and efficiently.
Also Read:
Samsung Expands Focus on Tier 2 and 3 Cities Ahead of Festive Season
