Modal Raises $750M at $15.75B Valuation to Power AI Agents

Modal, a cloud infrastructure startup building tools for AI applications and agents, has raised $750 million in a new funding round at a $15.75 billion valuation. The round was led by Accel, with Kleiner Perkins joining and all existing investors participating, including Menlo Ventures, General Catalyst, Redpoint, Lux Capital, and Bain Capital Ventures.

Announced on October 14, 2026, the funding marks a sharp increase from Modal's $4.65 billion valuation in its Series C round on May 21. The latest valuation is more than three times its previous level in less than five months.

Founded in 2021 by Erik Bernhardsson, Modal provides cloud infrastructure that allows developers to run code, train AI models, and deploy AI applications without managing the underlying servers. Its platform is increasingly being used to support AI agents that need computing resources to execute tasks, test generated code, and run applications.

The company reports that its annualised revenue run rate has crossed $1 billion. It also says usage has tripled since its Series C and grown more than 15 times year over year. Its Sandboxes product alone has reached a $300 million annualised run rate, making it a major contributor to the business.

Modal's $750 Million Funding Round and Rapid Valuation Growth

The $750 million round was led by Accel, with Kleiner Perkins participating alongside Modal's existing investors. The company did not disclose a detailed allocation of the funding among the participating investors.

The latest financing follows a series of rapidly increasing valuations. Modal was valued at $1.1 billion in September 2025 when it raised an $87 million Series B led by Lux Capital. It subsequently raised a $355 million Series C in May 2026 at a $4.65 billion valuation, led by General Catalyst and Redpoint.

The Series C reportedly involved two tranches, with the valuation increasing from $2.5 billion to $4.65 billion as additional investors entered the round. Accel and Menlo Ventures participated in the later tranche.

With the new Series D valuation of $15.75 billion, Modal's valuation has increased more than fourteenfold from its Series B level in just over a year.

Across its disclosed funding rounds, including an earlier seed round led by Amplify Partners and a $16 million Series A in October 2023, Modal has raised approximately $1.2 billion.

The scale of the latest round reflects investor interest in the infrastructure needed to support the growing use of AI models and autonomous agents. However, valuation growth does not necessarily translate into equivalent growth in revenue or profitability, and Modal has not disclosed its valuation methodology or profitability in the supplied report.

How Modal's Cloud Infrastructure Works

Modal was founded by Erik Bernhardsson, who previously worked on Spotify's music recommendation systems and spent six years as chief technology officer at Better.com.

His experience helped shape the company's focus on making cloud computing easier for developers. Running AI applications can involve complex infrastructure requirements, particularly when workloads need access to graphics processing units (GPUs), large models, and rapidly scalable computing capacity.

Modal brings computing capacity from more than 20 cloud providers into a unified network. Developers can run code on this infrastructure without having to manage individual servers themselves.

To support this approach, the company has developed its own file system, container runtime, scheduler, and networking infrastructure.

These components help developers execute workloads, allocate computing resources, and manage applications through a more unified development experience.

The underlying challenge is becoming more significant as AI applications move beyond generating text or answering questions. AI agents increasingly need to execute code, interact with software systems, run tests, process data, and complete tasks in the background. Each of these activities can require computing resources that must be provisioned and managed reliably.

Modal aims to provide the infrastructure layer that allows developers to build these applications without spending as much time managing cloud operations.

Sandboxes: A Major Growth Driver

One of Modal's most important products is Sandboxes, which launched in 2023 as developers increasingly needed environments in which AI-generated code could run safely.

A sandbox is an isolated computing environment where code can execute without directly interacting with a customer's main systems. For example, a coding assistant can generate a script and run it inside a sandbox to test whether it works before the code is used elsewhere.

This capability is particularly relevant for AI agents because generated code may need to be executed, tested, corrected, and executed again as part of a larger workflow.

Modal reports that Sandboxes has reached a $300 million annualised run rate. That represents just under one-third of the company's reported $1 billion-plus annualised run rate.

The company also reports that close to three billion sandboxes have been launched, with infrastructure capable of supporting up to one million running simultaneously.

These figures indicate the scale at which the product is being used, according to the company. They also highlight how a product originally launched to support code execution has become a major part of Modal's business.

As AI agents become more capable of carrying out multi-step tasks, demand for isolated environments could increase. Agents may need to run scripts, test software, process files, or execute different tasks concurrently.

However, sustained growth will depend on how much customers continue to use the service, what they are willing to pay, and whether competing cloud providers and AI platforms offer comparable capabilities.

Customers Across AI, Software, and Robotics

Modal serves companies using AI infrastructure for a range of applications, from software development to video generation and robotics.

Its reported customers include DoorDash, Cognition, Ramp, Linear, and Hebbia. The article says these companies use Modal for workloads involving background AI agents and application development.

DoorDash's AI platform team uses the infrastructure to move applications from experimentation into production.

Other customers use the platform for different computing-intensive activities. Runway generates video on Modal, while Physical Intelligence, Generalist AI, and 1X use it for robotics-related training workloads.

The company also highlighted TypeSafe AI's Jev product, which reportedly served one trillion tokens on Modal in its first three days. Tokens are units of text processed by AI models, and the figure illustrates the volume of model activity associated with the launch, according to Modal.

These use cases demonstrate the breadth of workloads that cloud infrastructure providers may support as AI adoption expands. The requirements differ between applications, but they often involve scalable compute, reliable execution environments, and the ability to move workloads from development into production.

Modal's strategy is to offer infrastructure building blocks that developers can combine according to their needs, rather than focusing exclusively on a single AI application.

A Growing Team Across Three Cities

Modal has more than 200 employees across New York, San Francisco, and Stockholm, according to the report. That represents an increase from more than 120 employees around the time of its Series C.

Bernhardsson serves as chief executive officer. Co-founder Akshat Bubna, who previously worked as an early engineer at Scale AI, serves as chief technology officer.

Bernhardsson also created Annoy, an open-source library used for approximate nearest-neighbour search and recommendation-related applications.

The founding team's experience across software infrastructure, AI, and large-scale technology products has informed Modal's focus on developer tools.

The latest funding provides capital for the company to continue expanding its infrastructure and supporting the growing range of workloads running on its platform. The company has not disclosed a detailed breakdown of how the $750 million will be allocated.

AI Infrastructure Spending Is Accelerating

Modal's funding arrives as demand for infrastructure to run AI models increases.

The report cites Gartner's forecast that worldwide spending on AI-optimised infrastructure-as-a-service will reach $42 billion in 2026, representing 96% growth. Of that total, inference spending, which covers running trained AI models, is forecast to reach $23.3 billion, compared with $19 billion for training.

Inference is especially relevant to AI agents because these systems repeatedly call models while performing tasks. They may also need computing resources to execute generated code, access tools, or operate software applications.

As businesses deploy more AI features into everyday workflows, infrastructure providers face pressure to make these operations faster, more reliable, and easier to scale.

Modal's focus on serverless execution and isolated sandboxes places it within this broader market for AI computing infrastructure.

The opportunity is substantial, but so is the competition. Customers can choose from multiple infrastructure providers, and large cloud platforms have the resources to develop their own services or integrate similar capabilities into existing products.

Competition From Baseten, Together AI, and Fireworks AI

Modal operates in a competitive market that includes companies focused on model inference, AI compute, and infrastructure for production workloads.

Baseten raised $300 million at a $5 billion valuation in January 2026, according to the article. It subsequently announced a reported $1.5 billion Series F at a $13 billion valuation in June.

Together AI raised $800 million at an $8.3 billion valuation on July 1, with the round led by Aramco Ventures. The company reported that its annual bookings had exceeded $1.15 billion in the preceding quarter.

Fireworks AI raised $1.505 billion at a $17.5 billion valuation on July 16, according to the supplied report, and reported a run rate above $1 billion.

These companies overlap with Modal in the broader AI infrastructure market, but their business models and technical focus are not identical. Together AI and Fireworks AI place particular emphasis on running and customising AI models, while Modal offers developers infrastructure components spanning functions, sandboxes, and training workloads.

Comparisons between the companies' financial metrics also require care. Annualised revenue run rate and annual bookings are different measures and should not be treated as directly interchangeable.

The rapid financing activity across these businesses nevertheless shows that investors are placing substantial capital behind infrastructure expected to support enterprise AI adoption.

Can Sandboxes Sustain Modal's Valuation?

Modal's latest funding highlights how quickly investor expectations for AI infrastructure businesses are changing. Its valuation rose from $4.65 billion in May to $15.75 billion in October 2026, while the company reported annualised run rate above $1 billion.

Sandboxes may prove central to whether the company can sustain that growth. The product already accounts for approximately $300 million in annualised run rate, according to Modal, and is designed for a use case closely linked to AI agents: running generated code in isolated environments.

The long-term opportunity depends on whether AI agents continue to generate workloads that require external execution infrastructure. If businesses increasingly rely on agents to write, test, and run code, demand for sandbox environments could expand.

At the same time, large cloud providers and AI model developers could introduce competing capabilities, potentially putting pressure on pricing and differentiation.

Modal's challenge is to turn its current growth into durable customer demand while maintaining the developer experience that helped attract its users.

For now, the company's $750 million funding round places it among the most highly valued players in AI infrastructure. Its next test will be whether its fast-growing platform, especially Sandboxes, can justify investor expectations as competition intensifies.