Nvidia has taken the next big step in the evolution of AI with its new Nemotron Coalition – a coalition that involves some of the world’s leading AI labs to work on the future of open frontier-level foundation models. The coalition is said to advance the cause of AI through shared research, expertise, data, and compute resources.
Announced during Nvidia’s GPU Technology Conference (GTC) in San Jose, California, the coalition brings together prominent players including India’s Sarvam AI, Mistral AI (France), Thinking Machines Lab (founded by former OpenAI CTO Mira Murati), Perplexity, Cursor, Reflection AI, Black Forest Labs, and LangChain as inaugural members.
Nvidia’s Nemotron Coalition co-developed base model on DGX Cloud
The coalition’s inaugural effort is a powerful base model jointly developed by Nvidia and Mistral AI, trained on Nvidia’s DGX Cloud infrastructure. This model will serve as an open foundation, enabling post-training customisation and specialisation for industries, regions, and specific applications. It will also underpin Nvidia’s upcoming Nemotron 4 family of models.
Nvidia founder and CEO Jensen Huang shared his take on the coalition, stating, “Open models are the lifeblood of innovation and the engine of global participation in the AI revolution—for students, scientists, startups and entire industries.”
Diverse expertise for inclusive and sovereign AI
As part of the coalition, each member contributes unique strengths to the coalition:
Sarvam AI: It brings expertise in sovereign, language-inclusive models, focusing on voice-first AI that understands local cultures and supports India’s multilingual needs.
Mistral AI: It provides frontier model development know-how, including efficient, customisable systems with full user control.
Thinking Machines Lab: Ex-OpenAI CTO Mira Murati’s TML offers data collaboration and a commitment to adaptable, broadly accessible AI.
Other participants add capabilities in high-performance systems (Perplexity), real-world evaluations (Cursor), multimodal tech (Black Forest Labs), agentic reasoning (LangChain), and open systems (Reflection AI).
Sarvam co-founder Pratyush Kumar highlighted the inclusive angle, stating, “AI reaches its full potential when it works in every language and for every community… Sarvam will contribute our support to the Nvidia Nemotron Coalition to build open foundation models that are voice-first, language inclusive, understand local culture and provide a platform for developers to build applications that matter at population scale.”
Mistral AI co-founder Arthur Mensch added, “Open frontier models are how AI becomes a ‘true platform’… Together with Nvidia, we will take a leading role in training and advancing frontier models at scale.”
Why Nvidia’s nemotron coalition is a big deal for open AI models
The Nemotron Coalition aims to push for transparency, collaboration, and sovereignty in AI development, aiming to broaden access to cutting-edge intelligence beyond closed ecosystems. By pooling resources, the group seeks to make frontier model creation more efficient, foster innovation, and ensure that AI advances benefit diverse global communities.
This initiative comes on the sidelines of Nvidia’s deepening ties with Indian AI efforts. Sarvam had previously leveraged Nvidia tools for its Indic-language models. The coalition also follows Nvidia’s strategic partnerships, including a major compute deal with Thinking Machines Lab.
As open-source AI gains momentum, the coalition positions Nvidia as a central part of a collaborative and democratised frontier intelligence, potentially reshaping how next-generation models are built and deployed worldwide.