NVIDIA invests $3 billion in Meta and expands work on Muse AI agents
NVIDIA will invest $3 billion in Meta's new equity offering. The companies are also expanding their work on Muse models, and Meta plans to deploy more NVIDIA Vera Rubin hardware.


NVIDIA will invest $3 billion in Meta's new equity offering. At the same time, the two companies are expanding their work on Meta's Muse models and AI agents, and Meta plans to deploy more of NVIDIA's next-generation data center hardware. The customer and the supplier are tying themselves closer together, both technically and financially.
NVIDIA released the announcement early on Tuesday, October 27, European time; its newsroom account shared it on X at 06:02 UTC. The day before, Meta launched a financing plan of about $95 billion to pay for its AI build-out.
What NVIDIA and Meta agreed
According to NVIDIA's press release, the companies will "partner to expand their work on Muse and the Muse family of models to improve inference and training performance." The aim is to optimize workflows for performance and lower total cost of ownership.
Meta intends to expand its deployment of NVIDIA infrastructure, including Vera Rubin systems and Vera CPUs for what the companies call next-generation AI agents. Meta will also deploy Spectrum-X Ethernet networking and BlueField-4.
Two parts of the deal are still exploratory. Meta and NVIDIA will "consider" further collaboration on NVIDIA's OpenShell Reference Architecture, which runs AI agents inside a contained, policy-controlled environment, and on NVIDIA Sentry, which monitors agent behavior. Meta has not committed to deploying either.
The press release does not say how many systems Meta will buy, what the hardware costs or when deliveries take place.
$3 billion into a $35 billion offering
The last sentence of NVIDIA's release is the headline: "NVIDIA will invest $3 billion in Meta's equity offering." In its forward-looking statements, NVIDIA describes the investment as proposed, subject to anticipated completion and any applicable closing conditions.
Meta's own release from Monday provides the context. The company launched two public offerings totaling about $35 billion: roughly $22.5 billion of Class A common stock and roughly $12.5 billion of mandatory convertible preferred shares. Meta also plans an at-the-market program of about $60 billion once its new Class C shares begin trading. Meta says the timing, size and terms of that program have not been determined.
The same release says NVIDIA, AMD and Broadcom have, severally and not jointly, indicated interest in buying up to $5 billion of Class A shares in aggregate, on the same terms as other buyers. Meta stresses that these indications are not binding. NVIDIA's announcement on Tuesday is the first concrete figure from any of the three chipmakers.
Why Meta is raising money
Meta expects capital expenditures of $135 to $145 billion in 2026, according to the release. The company says its AI investments are already generating "tangible returns" through higher engagement, better advertising performance and increased monetization. The money will fund infrastructure and compute for what Meta calls personal superintelligence, consumer and business agents, enterprise products and AI glasses.
Meta generated $97 billion of operating cash flow in the first three quarters and held about $82 billion in cash, cash equivalents and marketable securities at the end of the third quarter. It describes the offerings as a complementary source of capital alongside operating cash flow, debt and strategic partnerships.
NVIDIA CEO Jensen Huang calls AI agents "a new computing paradigm." Meta CEO Mark Zuckerberg says Muse is "how we bring personal superintelligence to everyone."
What business leaders should note
This is hogby.ai's assessment, not the companies' own claims.
- The money moves in a circle. NVIDIA is investing in one of its largest customers, which will at the same time deploy more NVIDIA hardware. Links like this make the AI market more fragile if growth slows, and belong in any risk assessment of a single supplier chain.
- Agent security is becoming infrastructure. Meta weighing OpenShell and Sentry shows that control over what AI agents do is moving down into hardware. Ask your own vendors how agents are isolated and monitored.
- A few players set the price of AI. When the largest buyers and suppliers bind themselves together, everyone else has less bargaining power over compute.
Sources and media
- NVIDIA: NVIDIA and Meta to Expand Collaboration on Muse and Muse Family of Models Amid Robust Growth
- NVIDIA Newsroom on X
- Meta: Meta Announces Launch of Offering of Depositary Shares and Offering of Class A Common Stock
Source: NVIDIA and Meta. NVIDIA's press release is dated October 27, 2026 and was shared on X at 06:02 UTC. Meta's release is dated October 26, 2026. Both were read by hogby.ai on October 27, 2026.
Illustration: AI-generated with Grok Bot / hogby.ai