Meta is rolling out more than one gigawatt of custom silicon developed with Broadcom, plus a significant amount of AMD chips, alongside new NVIDIA systems.

Judge as of 2026-04-29. The claim has three parts: (a) Meta is deploying more than 1 gigawatt of its own custom silicon (MTIA or similar) co-developed with Broadcom; (b) Meta is deploying a significant amount of AMD chips; (c) Meta is deploying new NVIDIA systems. 'Rolling out' means deployment underway or contracted and starting at that date. True: dated evidence (Meta, Broadcom, AMD or NVIDIA announcements, filings or credible reports) supports all three, with the Broadcom custom-silicon deployment sized at more than 1 GW. Mostly true: all three are supported but the Broadcom capacity figure is stated only as about 1 GW, or as a multi-year plan, not as a rollout already underway. Mixed: two of the three parts hold, or the 1 GW figure is clearly unsupported while the AMD and NVIDIA parts hold. Mostly false: only one part holds. False: evidence shows Meta was not deploying custom Broadcom silicon at anything close to that scale and was not deploying AMD chips in significant amounts.

Set before the research began.

1. E1 Meta Partners With Broadcom to Co-Develop Custom AI Silicon Meta (about.fb.com) · 14 April 2026 Own record

   “The agreement includes a commitment that exceeds 1GW, which is the first phase of a sustained, multi-gigawatt rollout, reinforcing a shared roadmap to co-design and scale the hardware required to bring personal superintelligence to billions of people globally.”

2. E2 Meta Partners With Broadcom to Co-Develop Custom AI Silicon Meta (about.fb.com) · 14 April 2026 Own record

   “As we roll out more than 1GW of our custom silicon to start and then multiple gigawatts over time, this partnership will give us greater performance and efficiency for everything we’re building.”

3. E3 Meta and AMD Partner for Longterm AI Infrastructure Agreement Meta (about.fb.com) · 24 February 2026 Own record

   “Today, we’re announcing a multi-year agreement with AMD to power our AI infrastructure with up to 6GW of AMD Instinct GPUs, the silicon computing technology used to support modern AI models.”

4. E4 Meta Builds AI Infrastructure With NVIDIA NVIDIA Corporation (GlobeNewswire release on investor.nvidia.com) · 17 February 2026

   “Meta will deploy industry-leading NVIDIA GB300-based systems and create a unified architecture that spans on-premises data centers and *NVIDIA Cloud Partner* deployments to simplify operations while maximizing performance and scalability.”

Meta's official Broadcom announcement (E1, E2), dated 2026-04-14, describes a commitment above 1GW of co-developed custom silicon. It calls this the first phase of a multi-gigawatt rollout. Meta's announcement uses the present-tense phrase "as we roll out" [E2], but neither quote independently confirms that deployment had already started. Meta's February 2026 announcement (E3) describes a multi-year agreement for up to 6GW of AMD Instinct GPUs, which supports the claim of a significant amount of AMD chips. NVIDIA's own release (E4), a source independent of Meta, says Meta will deploy GB300-based systems. All three parts are supported, but the Broadcom and AMD parts rest on Meta's own announcements of contracted plans rather than independent proof that deployment had started, so the claim is rated mostly true rather than true.