Nvidia (NASDAQ: NVDA) is the closest thing to a "picks and shovels" stock for humanoid robots. It sells the onboard computer (Jetson Thor), the foundation models (Isaac GR00T), the world-simulation models (Cosmos) and the training GPUs that most robot makers use, and it has invested in Figure AI. The catch for investors: robotics is a rounding error in a company that booked $96.2 billion of revenue last quarter, and since 2026 Nvidia no longer even reports robotics as its own line. This page measures how much humanoids really matter to NVDA. For the full list of humanoid plays, see our Tesla Optimus stocks hub and humanoid robot stocks guide.
- Ticker: NVDA, Nasdaq. Closed at $181.60 on Nov 5, 2026 (market cap ~$4.38T).
- Robotics share of sales: about 1% (automotive and robotics: $2.3B of $215.9B in FY2026).
- Reporting change: robotics now sits inside an "Edge Computing" segment ($7.2B in Q2 FY2027).
- Next catalyst: Q3 FY2027 results on Nov 17, 2026; guidance $108B revenue.
- Verdict: a data center AI stock with a robotics option attached, not a humanoid pure-play.
NVDA Snapshot
| Item | Value |
|---|---|
| Ticker / exchange | NVDA · Nasdaq |
| Recent close | $181.60 (Nov 5, 2026), down 9.3% that day |
| Market value | ~$4.38 trillion |
| 52-week range | $164.27 to $243.37 |
| Last reported quarter | Q2 FY2027 (ended Jul 26, 2026) |
| Next report | Nov 17, 2026 (Q3 FY2027) |
Price data from Gotrade, which tied the November 5 drop to a broad sell-off in AI and chip stocks rather than company news. Check a live quote before acting.
The Robotics Stack Nvidia Sells
| Product | What it does for a humanoid | Status |
|---|---|---|
| Jetson Thor (Jetson AGX Thor) | Blackwell-based onboard computer that runs the robot's AI models in real time | Generally available since Aug 25, 2025; developer kit priced at $3,499 |
| Isaac GR00T | Open foundation models for humanoid skills | New GR00T N models announced in 2026 |
| Cosmos | World models that generate synthetic training data | New Cosmos models released in fiscal 2026 and 2027 |
| Isaac Sim / Isaac Lab, Omniverse | Simulation to train and test robots before real-world use | New Isaac frameworks in 2026 |
| IGX Thor | Industrial-grade edge AI platform with functional safety | Generally available (Q1 FY2027) |
| Data center GPUs | Training robot models | The bulk of Nvidia's sales |
Sources: Nvidia's Q1 FY2027 release and SiliconANGLE on Jetson Thor. The strategy is to be the default toolkit whoever builds the robot, which is why Nvidia shows up in the investor lists of both Figure and other humanoid startups and in the parts lists of companies like Boston Dynamics.
How Big Is Robotics in Nvidia's Numbers?
| Period | Line that includes robotics | Revenue | Share of total |
|---|---|---|---|
| FY2026 (to Jan 2026) | Automotive and robotics | $2.3B (+39%) | ~1.1% of $215.9B |
| Q4 FY2026 | Automotive and robotics | $604M (+6%) | ~0.9% of $68.1B |
| Q1 FY2027 | Edge Computing | $6.4B (+29%) | ~7.8% of $81.6B |
| Q2 FY2027 | Edge Computing | $7.2B (+27%) | ~7.5% of $96.2B |
From fiscal 2027 Nvidia reports only two platforms: Data Center and Edge Computing. Edge Computing covers PCs, game consoles, workstations, AI-RAN base stations, robotics and automotive, per the Q1 release. Shares are our calculation. The last standalone figure comes from the fiscal 2026 results.
Our estimate: robotics on its own is likely still well under 2% of Nvidia's revenue. Even if humanoid computers grew tenfold, the data center business would still decide the stock. Investors buying NVDA "for robots" are mostly buying AI infrastructure.
Latest Results: Q2 Fiscal 2027
Reported on August 26, 2026, per Tradune's summary of the release. Non-GAAP EPS was $2.22. The Q3 guide assumes no data center compute revenue from China, and guided gross margin dips to 74.0%. Fiscal Q3 2027 results were not out at the time of writing.
What to Watch on November 17
- Edge Computing growth: the only reported line that captures robotics; a re-acceleration from +27% would hint at physical AI demand.
- Jetson Thor and robotics commentary: customer counts or volumes for humanoid makers, which Nvidia rarely quantifies.
- Data center guidance: after the early-November sell-off, this will drive the stock far more than robots.
Who Builds Humanoids on Nvidia
In its fiscal 2026 results Nvidia named Boston Dynamics, Caterpillar, Franka Robotics, the startup Humanoid, LG Electronics and NEURA Robotics among companies using its robotics stack. Boston Dynamics sits inside Hyundai (see Hyundai and Boston Dynamics stock), and Apptronik announced plans in 2024 to integrate Nvidia's general-purpose humanoid model with Apollo, although its main AI partner today is Google DeepMind. Nvidia also invested in Figure's 2024 and 2025 rounds. Because so many rivals depend on the same toolkit, Nvidia is a natural core holding in humanoid ETFs; see our humanoid robot ETF comparison.
Nvidia and Tesla Optimus
Tesla is both a customer and a rival. It has trained its AI on Nvidia GPUs in its data centers, but it designs its own inference chips for cars and Optimus and is planning its own chip fabrication, covered in Terafab chips. So Optimus success does not automatically mean Nvidia robotics revenue; Nvidia's upside is larger with the many humanoid makers that do not build their own silicon. For what runs inside Optimus, see Optimus hardware specs.
Risks
- Concentration: about 92% of revenue came from Data Center in Q2 FY2027; an AI spending slowdown hits NVDA regardless of robots.
- Less disclosure: the new segments make it harder to track robotics.
- Custom silicon: Tesla and others build their own robot chips.
- China: guidance excludes China data center sales, and export rules keep changing.
This article is for information only and is not investment advice.
FAQ: Nvidia as a Robotics Stock
Is Nvidia a good way to invest in humanoid robots?
It is a broad, indirect way. Nvidia supplies robot computers (Jetson Thor), foundation models (Isaac GR00T, Cosmos) and simulation tools to many humanoid makers, so it can benefit whichever company wins. But robotics is a very small part of its revenue: automotive and robotics together were $2.3 billion of $215.9 billion in fiscal 2026, about 1%. NVDA is mainly a data center AI stock. This is not investment advice.
How much revenue does Nvidia make from robotics?
Nvidia no longer reports it separately. Its last standalone figure was automotive and robotics revenue of $2.3 billion for fiscal 2026 (up 39%). Since fiscal 2027 it reports robotics inside a new Edge Computing segment, which also includes PCs, gaming consoles, workstations and cars; Edge Computing revenue was $7.2 billion in Q2 fiscal 2027.
What is Nvidia Isaac GR00T?
Isaac GR00T is Nvidia's family of open foundation models for humanoid robots, used together with its Isaac simulation tools and Cosmos world models to train robot behaviors. Nvidia released new GR00T N models in 2026.
When does Nvidia report Q3 fiscal 2027 earnings?
Nvidia scheduled its third-quarter fiscal 2027 conference call for Tuesday, November 17, 2026, at 2 p.m. Pacific time. Its guidance calls for revenue of $108 billion, plus or minus 2%.
Does Tesla use Nvidia chips in Optimus?
Tesla designs its own AI inference chips for its cars and Optimus, while it has used Nvidia GPUs to train AI models in its data centers. Tesla is therefore both an Nvidia customer and a competitor in robot computing.
The Bottom Line
Nvidia is the most widely used supplier of robot brains, models and simulators, and it profits whether Figure, Boston Dynamics or a Chinese maker wins. But robotics was only about 1% of revenue when Nvidia last broke it out, and it is now buried inside a $7.2 billion Edge Computing segment next to gaming and PCs. Buy NVDA for AI infrastructure with a robotics option, not as a humanoid pure-play. We will update this page after the November 17 results.
Sources: Nvidia (Feb 25, 2026; May 20, 2026), Tradune (Q2 FY2027, Aug 26, 2026), Gotrade (Nov 6, 2026), SiliconANGLE (Aug 25, 2025), Figure, The Robot Report. Not investment advice. Last fact-check: November 10, 2026.
FOLLOW THE ROBOT ECONOMY
We follow the companies that supply Optimus and its rivals, from chips and AI models to actuators and magnets, and update these pages after each earnings report.
