NvidiaNVDA
Nvidia makes the chips that power most of today's artificial intelligence.
Past week: -5.84%
30-day price
Where the chart sits — description, not prediction
Between its 50-day ($210.03) and 200-day ($190.23) averages — a trend in transition. 30-day range $200.04–$235.74; currently in the lower third of that range. RSI(14) 35 — momentum weak.
Computed from daily closing prices (Yahoo Finance), June 23, 2026. Compare all markets →
What is Nvidia?
Nvidia designs graphics processing units (GPUs) — chips built to do many calculations at once. It does not manufacture them itself; it designs them and has foundries like TSMC build them, and it also sells the software and networking that go alongside.
Founded in 1993, Nvidia spent its first decade on PC gaming graphics. Today its data-center GPUs — the H100 (2022) and Blackwell family (2024) — are the standard hardware for training and running large AI models like the ones behind ChatGPT.
Two things made Nvidia central to AI: GPUs are ideal for the parallel math that training requires, and its CUDA software (2006) became the foundation almost every AI tool is built on, making it hard to switch away.
What has moved Nvidia
2023 — the ChatGPT demand wave
After ChatGPT launched, companies rushed to buy Nvidia's H100 GPUs; a blowout May 2023 forecast ($11B vs. $7.2B expected) sent the stock up 239% for the year, and Nvidia first crossed $1 trillion in market value.
Feb–June 2024 — $2T, $3T, and a 10-for-1 split
Record data-center sales pushed Nvidia past $2 trillion in February 2024 and $3 trillion in June 2024; a 10-for-1 stock split took effect June 10, 2024, and on June 18 it briefly became the world's most valuable company. The stock rose about 171% in 2024.
Jan 27, 2025 — the DeepSeek shock
A Chinese startup's claim of a top-tier AI model trained cheaply sent NVDA down 16.9% in one day, erasing about $589 billion — the largest single-day market-value loss in history. The stock recovered as cloud spending plans held firm.
Mid-2025 — Blackwell ramp and $4 trillion
Surging Blackwell GPU shipments drove revenue from $44 billion to $57 billion per quarter through 2025; on July 10, 2025 Nvidia became the first company to close above a $4 trillion market value, later reaching about $5 trillion in October 2025.
FY2026 — record revenue
Nvidia's fiscal year ending January 2026 produced about $215.9 billion in revenue, up 65%, nearly all from data-center GPUs, leaving it the world's most valuable public company in mid-2026.
Notable moments
From $1T to most valuable in about a year
Nvidia first crossed $1 trillion in May 2023 and, by June 2024, briefly passed Microsoft as the most valuable company in the world — a compression of milestones with no real precedent.
CUDA: an 18-year software moat
Nvidia released CUDA in 2006 as a niche tool; by 2023 millions of developers had built AI software on it, which is widely cited as the reason rivals struggle to displace Nvidia even with comparable chips.
Common questions
What does Nvidia actually sell?
Mainly GPUs — data-center chips that cloud and AI companies use to train and run AI models, and GeForce gaming chips for PCs. Data-center products are now the large majority of its revenue.
Why does AI need Nvidia's chips specifically?
Training AI repeats the same math on huge blocks of data, which GPUs do far faster than ordinary processors — and Nvidia's CUDA software, which most AI frameworks rely on, makes switching to rivals costly.
What was the DeepSeek selloff?
On January 27, 2025 a Chinese model claimed to rival top systems at a fraction of the cost; fearing weaker GPU demand, investors sent Nvidia down 16.9% (~$589B) in a day. It largely recovered as spending plans held.
How fast has revenue grown?
From about $27 billion (FY2023) to roughly $215.9 billion (FY2026) — an eightfold rise in three years, almost all from data-center GPUs. This is context, not investment advice.