Daytona Raises $60 Million at $1 Billion Valuation to Build Cloud for AI Agents

A $1 billion bet on the infrastructure behind AI agents

Daytona announced a $60 million Series B at a $1 billion valuation on Oct. 20, 2026, led by FirstMark Growth, according to the company’s announcement and Fortune. The company said the financing brought its total funding to $91 million. TechCrunch reported on the round on Oct. 23, naming the same lead investor and valuation. ^{[1] [2] [3]}

The investor group includes technology founders Patrick Collison of Stripe, Tobi Lütke of Shopify, Aravind Srinivas of Perplexity, Eric Yuan of Zoom, Olivier Pomel of Datadog, Adam D’Angelo of Quora and Brendan Foody of Mercor, Daytona said. Qualcomm Ventures, Samsung Next and Okta Ventures also participated. Fortune reported that about $50 million of the round came from existing investors, a detail that shows how much of the financing was follow-on capital rather than new money. ^{[1] [2]}

What Daytona’s agent cloud does

Daytona is focused on the execution layer of AI software: the computing environment where an agent can run code, use a browser or interact with other tools. That is distinct from model training and inference, the computing work used to develop a model and generate its responses. The company describes its product as an agent execution cloud made up of computers, browsers and storage for software agents. ^{[1] [2]}

The company says its sandboxes can launch in milliseconds, retain state between tasks, be forked into parallel copies and then be discarded when work is complete. It says a single platform supports Linux, Windows, macOS and GPU environments. Fortune reported CEO Ivan Burazin’s estimate that a new sandbox can be configured in about 60 milliseconds; that is a company-reported performance figure, not an independently benchmarked comparison in the cited coverage. ^{[1] [2]}

Daytona lists AI model developers including Meta and Cohere among its customers, as well as companies such as Writer, Clay, n8n, Postman and Mintlify that build or deploy AI-enabled products. The company said Writer provisioned 300,000 sandboxes in one month and that AfterQuery replaced an in-house Kubernetes cluster with Daytona. These customer and usage examples come from Daytona and have not been independently audited in the sources cited here. ^[1]

A growing U.S. market, with execution still to prove

The round reflects a U.S. venture bet that agents will need more than access to a language model: they may require persistent, isolated compute to complete multi-step work. Daytona says its revenue grew about 25-fold from January through September and reports average monthly net revenue retention of roughly 135%, meaning its existing customer base spent more over time. Those are company-reported indicators, not audited financial results, and the funding announcement does not disclose revenue, margins or how much of the business is recurring. ^{[1] [2]}

The investment also puts Daytona in a market where established cloud providers can offer virtual machines and where other AI infrastructure startups are building tools for agent execution. The company argues that rapid startup, preserved state and parallel copies are better suited to agent workloads than conventional cloud instances; whether that specialization becomes a durable business will depend on reliability, security, cost and sustained customer demand. Daytona said it plans to use the new capital primarily to expand sales and go-to-market staffing, with additional engineering hires. ^{[1] [2] [4]}