Key facts
- Fiscal 2026 company revenue US$215.9bn; Q4 FY2026 revenue US$68.1bn; Data Center Q4 US$62.3bn (NVIDIA Q4 FY2026 earnings, 25 Feb 2026).
- Ownership: Public (Nasdaq: NVDA).
Focus areas
About
NVIDIA is the accelerated-computing company whose GPUs and systems underpin most large-scale AI training and inference. Founder and chief executive Jensen Huang leads the Santa Clara-based public company (Nasdaq: NVDA). In February 2026 NVIDIA reported fourth-quarter revenue of US$68.1 billion and full fiscal 2026 revenue of US$215.9 billion, with Data Center revenue of US$62.3 billion in the quarter. On the enterprise AI software surface — NVIDIA AI Enterprise, NIM inference microservices, NeMo model customisation and related agent/blueprint tooling — enterprises deploy open and proprietary models with production support, portable microservices and lifecycle tools across cloud, data centre and edge, typically on NVIDIA-accelerated infrastructure. Hyperscaler partnerships including AWS and Microsoft Azure are key distribution paths for the same software stack.
How they use AI
NVIDIA packages production AI as software, not only silicon. NVIDIA NIM provides prebuilt, optimised inference microservices with standard APIs, validated runtimes (including TensorRT-LLM, vLLM and SGLang backends) and Kubernetes-friendly containers so teams can deploy foundation models for language, retrieval, speech, vision and safety with enterprise support. NVIDIA NeMo covers the model lifecycle — data curation, customisation, evaluation, RAG, guardrails and agent tooling — so organisations can adapt and govern models before serving them. NVIDIA AI Enterprise bundles application-layer frameworks and microservices with infrastructure orchestration under commercial support and SLAs, including Production Branch and Long-Term Support Branch release trains. NVIDIA Blueprints add reference agent workflows (for example multimodal RAG and digital humans) that compose NIM and NeMo components.
Products and services
- NVIDIA NIM — optimised inference microservices for foundation models with standard APIs and enterprise support
- NVIDIA NeMo — platform for building, customising, evaluating and guarding generative and agentic AI models
- NVIDIA AI Enterprise — commercial software suite combining application-layer AI tools with GPU infrastructure management
- NVIDIA Blueprints — reference agentic and generative AI workflows with partner microservices and Helm charts
- NVIDIA Nemotron — NVIDIA’s open reasoning / agentic model family offered via NIM (software surface, not full silicon catalogue)
Notable customers / partners
Key people
- Jensen Huang — Founder and Chief Executive Officer
