24.03.2026, Moscow.
Nvidia Corporation announced the creation of an alliance of leading AI laboratories worldwide to develop an open artificial intelligence model, which could delay the collapse of the financial “bubble” formed around the AI market, Rossa Primavera News Agency IT Desk notes.
On March 16, Nvidia reported that the Nemotron coalition includes text-to-image model developer Black Forest Labs, the creator of the AI assistant for programmers Cursor, the framework for agent-based systems LangChain, the developer of open language models Mistral AI, the AI-powered search engine Perplexity, the startup focused on open frontier models Reflection AI, the voice technology company Sarvam, and Thinking Machines Lab, a laboratory for customizable AI systems founded by Mira Murati.
The coalition aims to create a next-generation AI model, Nemotron 4, by combining the unique contributions of each participant. It is planned that the model will be capable of working with code, images, voice, and search, functioning as an agent and understanding dozens of languages. Importantly, once completed, the model is intended to be released as open source. This means anyone will be able to run it on their own devices, fine-tune it with their own data, and avoid the risk of data leaks through cloud-based systems.
If successful, the project could challenge the positions of OpenAI, Anthropic, and Google in the AI market, possibly even more than the emergence of the Chinese model DeepSeek-R1.
This raises the question: why has Nvidia begun acting against OpenAI, with which it has closely cooperated in recent years? Their partnership has shown signs of strain. Nvidia has postponed plans to invest $100 billion in OpenAI. In turn, OpenAI has reportedly been dissatisfied with the quality of its partner’s chips and has started seeking alternatives. Additionally, observers have noted a cyclical pattern in which Nvidia invests in OpenAI, which then uses the funds to purchase Nvidia chips.
At first glance, the creation of the open Nemotron 4 model may appear altruistic. However, Nvidia’s business logic here is straightforward: the more powerful and accessible open models become, the more companies will want to run them, and consequently, the greater the demand for Nvidia’s GPUs. This could delay the bursting of the AI bubble for quite some time. However, in that case, hopes for a reduction in GPU prices are unlikely to materialize.
Source: Rossa Primavera News Agency
