Editorial illustration for Meta secures millions of Nvidia AI chips as Nvidia begins selling own AI CPUs
Meta Builds Custom AI Chip to Challenge Nvidia Dominance
Meta secures millions of Nvidia AI chips as Nvidia begins selling own AI CPUs
Why does this matter now? Nvidia has just begun offering its own AI‑focused CPUs to external customers—a move it’s never made before. That shift opens a new supply channel for firms that need massive compute power to train and run large language models.
At the same time, Meta is busy designing its own in‑house silicon to handle similar workloads. The juxtaposition is striking: a tech giant that has spent years building custom AI chips is now turning to a rival’s freshly released processors. While the tech is impressive, the scale of the purchase hints at how quickly Meta wants to fill the gap between its internal projects and the market’s demand for hardware.
It also raises questions about how long Meta will rely on Nvidia’s new CPUs before its own designs can take over. The partnership signals a pragmatic approach to securing the compute needed for the next wave of AI development.
Meta's new deal with Nvidia buys up millions of AI chips.
Meta's new deal with Nvidia buys up millions of AI chips Nvidia's starting to sell AI CPUs for use by themselves for the first time. Nvidia's starting to sell AI CPUs for use by themselves for the first time. Meta is also working on its own in-house chips for running AI models, but according to the Financial Times, it has run into "technical challenges and rollout delays" with its chip strategy. Nvidia is also dealing with concerns about depreciation and chip-back loans used to finance AI buildouts, as well as the pressure of competition.
Will Meta’s data centers run smoother? The new multiyear agreement adds millions of Nvidia Grace CPUs and Blackwell and Rubin GPUs, marking the first large‑scale Grace‑only deployment. Nvidia claims the chips will bring “significant performance‑per‑watt improvements” for Meta’s workloads, yet independent verification is pending.
In addition, the contract schedules Vera CPUs for a 2027 rollout, expanding the hardware mix even further. Nvidia’s move into selling AI‑focused CPUs to external customers is itself a notable shift, though how this will affect pricing or supply dynamics remains unclear. Meta continues to develop its own in‑house AI processors, but the article offers no detail on timelines or integration plans, leaving the relationship between proprietary silicon and Nvidia’s offerings uncertain.
The partnership underscores Meta’s reliance on Nvidia’s ecosystem for its AI ambitions, while also hinting at future diversification. Whether the promised efficiency gains translate into measurable cost or performance benefits for Meta’s services is still to be confirmed.
Further Reading
- Nvidia and Meta expand GPU team up with millions of additional AI ... - AOL Finance
- Nvidia resets the economics of AI factories, again - SiliconANGLE - SiliconANGLE
- Nvidia AI Chips Face Rising Competition - InsiderFinance - InsiderFinance
- At CES, NVIDIA and AMD Made Memory the Future of AI - Futurum - Futurum Group
Related Reading
Browse all Meta news →
Business & Startups OpenAI, a Series F San Francisco startup founded in 2015 by eight pioneers
Business & Startups Terminal-Bench 2.0 launches with Harbor, testing any container-installable agent
Business & Startups Zuckerberg Unveils Meta Compute to Build Global AI Infrastructure
Related Topic Meta's DreamGym boosts AI agent success by 30% over baseline methods
LLMs & Generative AI
Related Topic Study finds condensed metaphors and rhythmic framing can evade safety filters
LLMs & Generative AI
From Archives Google Sets Dates for I/O 2026, Promises Keynotes, Demos and Fireside Chats
From Archives Qodo 2.1 links memory to agents, cutting coding errors by 11%
From Archives Meta plans facial recognition for AI smart glasses, amid privacy concerns
Open Source
From Archives Yann LeCun‑linked startup challenges LLM orthodoxy after leaving Meta
LLMs & Generative AI
Common Questions Answered
Why is the $100 billion Nvidia-OpenAI deal currently stalled?
[cnbc.com](https://www.cnbc.com/2026/02/03/nvidia-openai-stalled-on-their-mega-deal-ai-giants-need-each-other.html) reports that negotiations are 'on ice' after some within Nvidia expressed doubts about OpenAI's business model. Despite the friction, both companies still need each other, with Nvidia requiring customers like OpenAI to drive chip sales and OpenAI needing Nvidia's AI chips to meet growth targets.
What did Jensen Huang say about the proposed $100 billion OpenAI investment?
[fortune.com](http://fortune.com/2026/02/02/jensen-huang-nvidia-ceo-on-openai-investment-never-a-commitment/) quotes Huang stating that the $100 billion investment was 'never a commitment' and that Nvidia would consider funding rounds 'one at a time'. Huang also told CNBC that he is 'looking forward to Sam closing it' and will 'invest in the next round'.
What is the current valuation of Nvidia and OpenAI according to the article?
As of the article's publication, Nvidia's market cap peaked at $5 trillion in October and has since dropped to $4.4 trillion. OpenAI was valued at $500 billion on the private market late last year and is reportedly eyeing a valuation over $800 billion in its next funding round.