Meta’s approach to open-source artificial intelligence (AI) has been a major talking point in the tech industry. While the company has publicly emphasized that selling access to its Llama AI models is not part of its business strategy, a newly unredacted court filing suggests otherwise. The filing, part of the ongoing copyright lawsuit Kadrey v. Meta, reveals that Meta has revenue-sharing agreements in place with companies hosting its Llama models.
Llama’s Monetization Strategy Unveiled
In a blog post last July, Meta CEO Mark Zuckerberg asserted that “selling access” to Llama models was not Meta’s primary business model. However, legal documents now indicate that Meta does, in fact, generate revenue from Llama through partnerships with various hosting platforms. Specifically, the filing states that Meta receives a percentage of the revenue generated by companies that provide hosting services for Llama-based applications.
The filing does not specify which companies are compensating Meta. However, Meta has publicly listed several cloud and AI infrastructure providers as its Llama host partners, including AWS, Nvidia, Databricks, Groq, Dell, Microsoft Azure, Google Cloud, and Snowflake. These platforms offer developers the ability to access and deploy Llama models with enhanced scalability, performance optimization, and enterprise-grade tools.
While developers are not required to use these hosting partners to run Llama models, many choose to do so for the additional support and integrated services they provide. Llama models can be downloaded and fine-tuned independently, but leveraging cloud-based infrastructure simplifies the deployment process for businesses and researchers alike.
A Shift in Meta’s AI Strategy?
Zuckerberg had previously hinted at the possibility of monetizing Llama during an earnings call in April 2024. He suggested that Meta might explore revenue streams through business messaging services, AI-powered advertisements, and licensing agreements with major tech players. At the time, he stated:
“If you’re someone like Microsoft, Amazon, or Google and you’re going to be reselling these services, that’s something that we think we should get some portion of the revenue for. So those are the deals that we intend to be making, and we’ve started doing that a little bit.”
This aligns with the new court filing, which confirms that revenue-sharing arrangements are indeed in place. Despite this, Zuckerberg has continued to emphasize the broader benefits of an open-source approach, arguing that community contributions to Llama enhance Meta’s AI capabilities and strengthen its overall ecosystem.
During Meta’s Q3 2024 earnings call, he stated:
“I think it’s good business for us to do this in an open way. It makes our products better rather than if we were just on an island building a model that no one was kind of standardizing around in the industry.”
Legal and Financial Implications
The revelation of Meta’s revenue-sharing strategy is particularly significant given the legal battle it faces in Kadrey v. Meta. The plaintiffs in the lawsuit accuse Meta of using massive amounts of pirated e-books to train Llama models. They further allege that Meta engaged in “seeding” these copyrighted materials—effectively distributing them via torrenting methods. If proven true, this could pose serious legal and reputational risks for Meta.
Amid these legal challenges, Meta continues to double down on AI investments. The company has announced plans to increase its capital expenditures (CapEx) significantly in 2025, with an estimated budget of $60 billion to $80 billion—double its 2024 spending. These investments will primarily be directed toward expanding data centers and enhancing Meta’s AI research and development efforts.
To offset these rising costs, Meta is reportedly considering launching a paid subscription model for Meta AI, offering additional capabilities to its AI assistant. While details remain scarce, such a move could signal a broader shift toward AI monetization beyond infrastructure partnerships.
The Future of Llama and Meta’s AI Ambitions
Meta’s AI strategy is evolving rapidly, balancing open-source collaboration with monetization efforts. While Zuckerberg has positioned Llama as a tool for advancing AI research and innovation, the emergence of revenue-sharing agreements indicates that Meta is not entirely foregoing financial gains from its AI ventures. As the Kadrey v. Meta lawsuit progresses, more details about Llama’s development and Meta’s business strategies may come to light.
For now, one thing is clear: Llama is more than just an open-source initiative—it’s a critical component of Meta’s AI roadmap, and its financial and legal implications are only beginning to unfold.
