Wire flash
TechNvidia launches Spectrum-6 networking system; Tesla and SpaceX among early adopters
Editorial responsibility
- No named human review is recorded for this page.
- Source reporting is collected, normalized, translated or condensed automatically when needed.
- Automatically published source-backed update
Nvidia has unveiled Spectrum-6, a new networking system designed to connect hundreds of thousands of chips in AI data centers, doubling the capacity of its predecessor. The launch highlights a shift in the AI race from raw chip performance to efficient chip interconnection. Early adopters include cloud providers Microsoft, CoreWeave, and Nebius, as well as Tesla and SpaceX, which are building in-house AI systems for self-driving and robotics (Tesla) and xAI/Grok (SpaceX). Nvidia's Data Center revenue surged 92% year-over-year to $75 billion in Q1 FY2027, driving total revenue up 85% to $81.6 billion. Despite a 22% stock gain over the past year, NVDA trades at a forward P/E of 23.5, below its five-year average, suggesting the market may be undervaluing its expanding infrastructure role.
Source report
Author: Jabran Kundi Read Time: 5 min
Overview
When people discuss the artificial intelligence (AI) race, the conversation typically centers on chips. However, Nvidia's (NVDA) latest launch highlights a quieter but increasingly critical bottleneck: networking.
Nvidia has unveiled Spectrum-6, a new networking system designed to connect the massive number of chips inside modern AI data centers. The system doubles the capacity of its predecessor and is built for facilities running hundreds of thousands of chips simultaneously.
Why Networking Matters
As AI data centers grow larger, the challenge shifts from simply having enough chips to keeping them all working together efficiently. Key points:
- If even one connection is slow, every other chip must wait, leaving expensive hardware idle.
- Networking ties chips together and has become a major factor in AI factory performance.
- Spectrum-6 is Nvidia's solution to this problem, signaling that the company now sells far more than just chips.
Early Adopters: SpaceX and Tesla Stand Out
The list of early adopters includes cloud providers such as Microsoft (MSFT), CoreWeave (CRWV), and Nebius (NBIS). However, two names are particularly notable:
- Tesla (TSLA): Building AI systems for self-driving and robotics software.
- SpaceX (SPCX): Developing AI capabilities for its xAI and Grok initiatives.
Neither company rents out computing power commercially. Their inclusion on the list underscores the scale of their in-house AI ambitions and marks a shift in where the AI race is being fought.
The competition is no longer only about who has the fastest chips. It is also about who can tie those chips together most efficiently. That is where Nvidia is pulling ahead.
About Nvidia Stock
Nvidia operates as a leading AI infrastructure company, designing GPUs, networking solutions, and software for data centers, gaming, and automotive applications. Founded in 1993, the company is headquartered in Santa Clara, California.
Stock Performance
- NVDA traded sideways until March 2026, then grew rapidly in May, reaching a 52-week high of $236.54.
- Recently saw a slight dip on concerns over reported delays for next-generation chips.
- Over the past 12 months, the stock is up 22%, underperforming the iShares Semiconductor ETF (SOXX), which gained 128% over the same period.
- Despite this, NVDA has outperformed the S&P 500 Index ($SPX) over the past year.
Valuation
- Market capitalization: Over $5 trillion, making Nvidia one of the world's most valuable companies.
- Forward price-to-earnings (P/E) ratio: 23.5x — well below the five-year average.
- Forward price-to-sales (P/S) ratio: 23.2x.
- Earnings are forecast to grow 92% in fiscal 2027.
- The balance sheet further supports an attractive valuation.
Data Center Business Drives 85% Revenue Jump
Nvidia reported its first-quarter fiscal 2027 earnings on May 20:
| Metric | Value | Year-over-Year Change | |--------|-------|-----------------------| | Total Revenue | $81.6 billion | +85% | | GAAP Gross Margin | 74.9% | Largely flat sequentially | | Data Center Revenue | $75 billion | +92% | | Hyperscaler Revenue (Data Center) | $38 billion | +100% |
Hyperscalers accounted for more than half of all data center revenue.
Key Takeaway
Launches like Spectrum-6 demonstrate Nvidia's expansion beyond chips into networking — the infrastructure that ties AI factories together. The more of this system Nvidia controls, the more its customers will depend on it. While the market has priced NVDA stock as if its best days are behind it, the company's product roadmap suggests otherwise.
This article originally appeared on Barchart.
Source
Yahoo FinanceWestern
Part of this Story
Nvidia Launches Spectrum-6 Networking System; Tesla and SpaceX Among Early Adopters