Modal Labs, a New York–based startup that helps developers run inference workloads without owning servers, is closing in on a $750 million funding round at a $15.75 billion valuation. This jump more than triples its worth since its last raise just four months ago. It comes as demand surges for inference infrastructure—the part of AI systems that generates output from already trained models.
What’s Changing
The pending financing, which appears to be led by Accel, represents a major leap for Modal, whose previous funding round valued it at around $4.65 billion. The company declined to comment on the deal, which has only been partially reported until now.
Other inference-focused startups are also seeing rapidly escalating valuations. One firm in discussions with investors could reach around $26 billion, roughly double its worth from June. Companies serving image and video generation workloads are similarly in advanced talks for funding. Although many of these players are hitting $1 billion in annualized revenue or have a reasonable shot by year end, profitability remains elusive. The cost of compute continues to put heavy pressure on margins.
Modal’s Position & Earlier Hurdles
Since launching in 2021 by Erik Bernhardsson and Akshat Bubna, Modal Labs has built a platform to enable developers to train and run heavy compute workloads without managing underlying infrastructure. Its clients include a range of AI-native and fintech companies. By May, Modal had reached over $300 million in annualized revenue.
Modal’s operations were recently in the spotlight due to a security incident: a customer mistakenly left an unauthenticated endpoint exposed, which a rogue agent used to run code inside the customer’s sandbox. Modal itself said its systems were not breached. The issue stemmed from the customer’s misconfiguration, not a flaw in Modal’s own platform.
Why It Matters
Inference—turning models into usable applications—is growing into a vital layer of the AI stack. As large language models and generative systems continue expanding, so does the need for reliable, scalable inference providers. Modal Labs and rivals are seeing sky-high interest from customers who want to avoid building data centers themselves.
Still, the scaling pains are real. Compute leases, hardware costs, and infrastructure complexity eat into margins even as revenue rises. For Modal, investors are betting those challenges can be overcome. If the round closes at these terms, it would underscore how hot the sector has become. The rise in valuation—and risk—points to both opportunity and reckoning ahead.
Analytically, this signals a turning point in how venture is flowing in AI infrastructure. Modal’s leap reflects not only confidence in its tech and revenue growth but also in the broader category of inference providers. Investors are increasingly assigning multibillion-dollar values to companies that can deliver inference at scale. That puts pressure on these firms to march toward operational efficiency, margin improvement, and security resilience. Watch how Modal allocates this capital: scaling hardware, optimizing costs, or expanding into global infrastructure may separate winners from the rest.
