If vLLM already solved LLM serving, why did SGLang appear?
Summary
Neysa Accelerates India’s Sovereign AI Mission, Delivering Unprecedented Control and Choice to the Nation’s AI Builders
Mumbai: Neysa, a provider of AI Acceleration Cloud Systems, today announced a major strategic move to give Indian organizations full control over their AI, ending the era of ‘black box’ dependency on foreign-hosted platforms. The addition of OpenAI’s GPT-OSS models to its extensive open-weights catalog on Velocis, Neysa’s AI-native cloud platform, serves as powerful proof of this mission, demonstrating that the world’s most advanced models can now be hosted, scaled, and secured entirely within India’s borders.
Until now, AI adoption in India has been hampered by opaque, overseas-hosted models that constrained innovation, complicated compliance, and created unpredictable costs. Neysa addresses this head-on by expanding its catalog to include GPT-OSS alongside leading families such as Llama, DeepSeek, Qwen, and Mistral. Models from all of these leading open-weights publishers are now available on Neysa’s Velocis Cloud as Managed Inference Endpoints.
Speaking on the strategic importance of this initiative, Sharad Sanghi, Co-founder and CEO of Neysa, said, “India cannot afford to build its AI future on platforms it does not control. The size of our digital economy, the sensitivity of our data, and the urgency of innovation demand indigenous infrastructure. By bringing world-class models like GPT-OSS into our Velocis ecosystem, we are proving that advanced AI can be hosted and scaled entirely within India. This gives enterprises the freedom to innovate without external permission and policymakers the confidence that AI is aligned with Indian regulations. For our developers, it creates a new frontier where their IP is protected and their creativity fuels India’s competitive advantage. Sovereignty in AI is the ability to define our digital destiny, and Neysa is building the control plane that makes it possible.”
Adding a product and technology perspective, Karan Kirpalani, Chief Product Officer of Neysa, said, “The conversation in AI is no longer just about what the technology can do, but where and how it is controlled. With Velocis, we have built a platform that ends the dependency on closed models that can only be billed on tokens, by giving users full transparency, predictable economics, and the freedom to build on their own terms. What excites me most is our pace; we integrated models like GPT-OSS into our catalog within weeks of their release, ensuring Indian enterprises have access to the best global innovation while retaining local sovereignty. This speed, combined with secure operations and compliance readiness, gives India’s builders the confidence to truly own their AI journey.”
This catalog expansion empowers developers with immediate access to models like GPT-OSS-120B, a Mixture-of-Experts architecture for state-of-the-art reasoning, and GPT-OSS-20B, optimized for high-performance deployments. It also includes specialized, fine-tuned versions such as GPTQ, AutoRound, and 4-bit models for custom workloads. The Velocis platform accelerates the entire process, enabling rapid deployment from concept to production API in minutes, with managed GPU orchestration and a unified API to switch between models seamlessly.
With its expanding catalog of open-weight models now live on Velocis, Neysa is positioning India to lead in the age of Sovereign AI. This expansion is a cornerstone of Neysa’s long-term strategy to build a complete, end-to-end AI ecosystem in India. The company will continue to invest heavily in its AI-native infrastructure, forge deeper partnerships, and introduce new platform services designed to manage the entire AI lifecycle, from data to deployment, securely and efficiently.
