Modal Announces Completion of $355 Million Series C Financing, Post-Money Valuation of $4.65 Billion, Positioned as Native Cloud Infrastructure for the AI Era
Modal announced the completion of $355 million in Series C financing, with a post-money valuation of $4.65 billion, led by General Catalyst and Redpoint, with Menlo Ventures and Accel joining, and all existing shareholders participating in the round.
Since September last year, the company's revenue has grown fivefold, with an annualized revenue exceeding $300 million.
Modal is positioned as native cloud infrastructure for the AI era, providing core capabilities such as low-latency elastic inference, dynamic agent runtime, and reinforcement learning sandbox, serving leading teams like DoorDash, Cognition, Decagon, and Physical Intelligence.
Source: Public Information
ABAB AI Insight
Modal was founded by former Databricks and Google engineers. This round of financing continues its evolution from an AI computing platform to a full-stack infrastructure. Previously, it had deeply optimized GPU snapshotting to achieve a 100x acceleration in cold starts and built its own storage and global multi-data center capacity pool.
In terms of capital strategy, Modal will direct new funds towards low-latency inference, sandbox expansion, and reinforcement learning training loops, concentrating resources on its own computing layer and contributions from open-source engines (vLLM, SGLang). The motivation is to seize the wave of agent and model ownership, helping teams achieve full lifecycle control from fine-tuning to production, while forming a differentiated moat through the sandbox (which already accounts for over one-third of revenue).
Similar to the early high-valuation expansions of Snowflake and Databricks as AI data/computing layers, and competition from GPU cloud providers like Lambda and Together, the current AI infrastructure sector is transitioning from single GPU leasing to AI-native full-stack platforms. Early technology-driven teams are capturing mid-to-large AI company budgets through developer experience and flexibility.
Essentially, this represents capital concentration: Modal, as an AI-specific cloud, will shift pricing power from traditional general clouds (AWS/Azure) to platforms designed specifically for AI. The mechanism lies in the unique demands of AI workloads for elasticity, isolation, secure sandboxes, and training-inference loops, which traditional cloud architectures cannot efficiently adapt to, leading to rapid capital concentration towards specialized players like Modal and forming a new infrastructure tier.
ABAB News · Cognitive Law
The more the cloud is born for AI, the more developers are willing to entrust the entire lifecycle to it.
When sandbox revenue accounts for one-third, agents are no longer a future prospect but a current revenue engine.
Traditional clouds are built for the web, while AI clouds are built for agents; whoever builds the right infrastructure first will reap the next generation of computing dividends.