Overview
- Nvidia reported record fiscal Q2 revenue of $96.2 billion, driven by $89.0 billion in Data Center sales that reflect rapid customer spending on large AI models and services.
- The company said Vera Rubin entered full production and is operating at CoreWeave, Google Cloud, Microsoft Azure, Oracle Cloud Infrastructure and Nebius, signaling broader commercial use of its AI factory platform.
- Supply and capacity commitments jumped to $279 billion and management disclosed about $108.5 billion of land, power and shell guarantees that back customer data center builds, including arrangements tied to the PORTS‑Pike site leased to OpenAI.
- Nvidia told investors it generated roughly $20 billion in CPU revenue in fiscal 2027 and expects to more than double that by fiscal 2028 as it pushes integrated CPU/GPU stacks to serve agentic AI workloads.
- Management guided around $108 billion for fiscal Q3, flagged a near‑term gross‑margin dip to about 74% because of higher memory and build costs, and said third‑party financing deals with major asset managers aim to scale the infrastructure buildouts needed by customers.