Wall Street finished Tuesday with a split personality. The S&P 500 was almost unchanged and the Dow Jones Industrial Average declined, yet the Nasdaq Composite gained 0.45% and closed at a record. The strongest buying remained concentrated in artificial intelligence, semiconductors and the companies supplying the infrastructure and data needed to build AI systems.
Astera Labs (NASDAQ: ALAB) rose 6.6% to $363.46, extending its gain across the last two sessions to nearly 20%. Innodata (NASDAQ: INOD) climbed 14.6% to $69.97.
- The rally was concentrated in AI. The Nasdaq reached a record while the S&P 500 was flat and the Dow declined.
- ALAB benefited from AI-infrastructure demand. Its Scorpio ramp and new Leo memory products strengthened growth expectations.
- INOD surged on a possible Meta connection. Its agent-training program may support Muse, but neither company has confirmed the link.
Why AI Stocks Led the Market
The wider market backdrop helped growth stocks. Brent crude fell about 2% to roughly $98.30 per barrel, while the 10-year Treasury yield eased to around 4.95%. Lower energy prices reduce near-term inflation pressure, and declining yields make the future earnings of fast-growing technology companies more valuable in present-day terms.
The more important driver was renewed confidence that AI spending can produce commercial results. Meta’s Muse assistant reached the top of Apple’s U.S. App Store after its September launch. Its early adoption encouraged investors to look beyond training costs and consider the potential revenue from subscriptions, transactions and enterprise use.
Successful agents also require more inference capacity, memory, connectivity and model evaluation, directly supporting the markets served by Astera Labs and Innodata.
Falling oil prices and hopes surrounding this week’s Trump-Xi meeting also improved risk appetite. Still, the market’s narrow performance showed that investors were not buying everything. They were favoring companies with direct exposure to AI adoption.
Why Astera Labs Rose 6.6%
There was no major new company announcement from Astera Labs on Tuesday. The move appears to have been a continuation of Monday’s semiconductor rally, when ALAB gained more than 12%, combined with growing interest in the company’s recent product announcements.
Astera Labs builds semiconductor-based connectivity products that help processors, memory and networking components exchange data across rack-scale AI clusters.
The company’s latest Leo memory-controller expansion gave investors a timely reason to revisit that opportunity. Astera said its new Leo X-Series is designed for memory-intensive agentic AI workloads.
In internal testing, the system delivered up to a 62% reduction in time to first token and up to 22% more tokens per second. The company also said the expanded product family is generating broader engagement and design wins among AI labs, hyperscalers and neocloud customers.
Those product claims are supported by strong reported growth. Astera generated second-quarter revenue of $392.4 million, up 104% from the prior year. Management guided for third-quarter revenue of $540 million to $560 million, with the Scorpio X-Series fabric-switch ramp expected to become a major growth driver.
Investors increasingly see Astera as a platform exposed to several AI bottlenecks, including PCIe connectivity, memory expansion, Ethernet signal conditioning and scale-up fabrics.
The risk is valuation. At Tuesday’s price, Astera’s market capitalization was approximately $63 billion, representing a substantial premium to its current revenue base. That leaves little room for production delays, slower hyperscaler spending or weaker-than-expected adoption.
Further upside will require the Scorpio ramp and new Leo design wins to become substantial revenue, not simply promising technology.
Why Innodata Jumped 14.6%
Innodata had a clearer same-day catalyst. Hunterbrook Media published research arguing that Innodata’s work on the personalization of long-horizon AI agents could be connected to Meta’s Muse assistant.
The report cited former employees, hiring activity and similarities between Innodata’s disclosed program and Meta’s description of Muse.
The possible connection attracted investors because Muse has become a prominent example of consumer demand for agentic AI. If Innodata is helping train or evaluate it, continued adoption could support demand for datasets, reinforcement-learning environments, testing and human feedback.
However, the link remains unconfirmed. Neither Meta nor Innodata has stated that Innodata worked on Muse. Hunterbrook also disclosed that its affiliated investment firm held a long position in INOD when the report was published. Investors should therefore separate the confirmed facts from the report’s interpretation.
The confirmed facts are still significant. In August, Innodata said a program with its largest customer covering the personalization of long-horizon agents was scaling. It also disclosed a second program involving reinforcement-learning environments for computer-use agents.
Second-quarter revenue increased 58% to $92.1 million, adjusted EBITDA reached $25.4 million and adjusted gross margin expanded to 49%. Management maintained guidance for full-year revenue growth of at least 40%.
The business is also diversifying, although customer concentration remains high. Its largest customer represented 37% of second-quarter revenue, while a second Big Tech customer accounted for 34%.
That diversification is an improvement from relying on a single account, but the top two customers still generated roughly 71% of quarterly sales.
At $69.97, Innodata was valued at approximately $2.43 billion with a trailing price-to-earnings ratio near 53. That valuation can work if agentic AI programs expand and margins remain strong. It could also contract quickly if a major customer reduces spending or if the reported Muse connection proves less material than investors currently expect.
What Investors Should Watch Next
For Astera Labs, investors should monitor the Scorpio production ramp, the conversion of Leo design wins into revenue and gross margins.
For Innodata, the key questions are whether its agent programs continue to scale and whether growth outside its two largest accounts accelerates. Confirmation of work connected to a successful consumer agent would be meaningful, but Tuesday’s report was not confirmation.
Tuesday’s session showed that investors are rewarding companies that solve practical AI bottlenecks around memory, connectivity, training data and agent evaluation.
Astera Labs and Innodata occupy different parts of that stack, which explains why both attracted strong buying even as the broader market barely moved.
The opportunity remains substantial, but expectations are already high. From here, sustained stock upside will depend on reported revenue, customer expansion and profitable execution rather than AI enthusiasm alone.
Disclaimer
The author may hold positions in one or more securities mentioned in this article. This material is provided for informational purposes only and does not constitute investment advice or a recommendation to buy or sell any security. Investors should conduct their own research and consider their financial objectives and risk tolerance before making investment decisions.
Marc has been involved in the Stock Market Media Industry for the last +5 years. After obtaining a college degree in engineering in France, he moved to Canada, where he created Money,eh?, a personal finance website.
