Hydra Host has raised $100 million in Series A financing to expand the global availability of GPU infrastructure for artificial intelligence workloads. The round was led by Kindred Ventures and included NVIDIA, ARK Invest, SPLY Capital, Era Funds, Comcast Ventures, Magnetar and PEAK6, alongside returning backers Founders Fund, 10x Founders, Sterling Road and Flume Ventures. The Boulder, Colorado-based company plans to use the capital to add GPU-as-a-Service capacity and extend the operational network that links data center owners with AI companies seeking dedicated compute.
Scaling AI Factory Capacity
The financing arrives as demand for specialised infrastructure continues to rise across model training, inference and research, placing greater emphasis on deploying and managing GPU fleets at scale. Hydra Host positions itself as an intermediary on both the supply and demand sides of this market, helping data center operators commercialise installed capacity while giving AI customers access to high-performance bare-metal hardware. Its strategy is centred on making infrastructure available through market-based rates while addressing security, data sovereignty and reliability requirements for mission-critical workloads.
A Software Layer for Data Centers
At the core of the company’s model is the Brokkr AI Factory Operating System, a platform designed to standardise the procurement, provisioning and orchestration of GPU resources. Hydra Host said the system provides operators with performance and revenue visibility, reducing the need to assemble separate proprietary software stacks that can complicate integrations with customers. The company said Brokkr is deployed across more than 50 data centers in the Americas, Asia-Pacific and Europe, the Middle East and Africa, where it supports the rollout and operation of AI-focused facilities.
Expanding the Compute Offtake Network
Alongside the operational software, Hydra Host runs a compute offtake network intended to match available capacity with demand from frontier labs, inference platforms, GPU marketplaces, neocloud providers and enterprise teams. The company said this network supports both short-term and longer-duration compute requirements, enabling customers to source infrastructure across different regions without having to build and maintain their own physical deployments. New funding is expected to support international expansion, procurement and supply-chain activity, 24-hour high-performance computing engineering, customer support and growth of the company’s software and go-to-market teams.
Partnerships and Market Position
Hydra Host has also highlighted partnerships and customer relationships as validation of its approach, including a GPU-as-a-Service proof of concept with Verizon Business that combined Verizon’s network with the Brokkr platform. Verizon Business Chief Product Officer Scott Lawrence said the trial demonstrated a route for converting network infrastructure into scalable AI capacity with an emphasis on speed, security and efficiency. Parasail Chief Executive Mike Henry said Hydra Host has supported the company’s GPU sourcing and the deployment of its initial in-house B300 fleet, adding that a broader infrastructure footprint could help meet regional inference demand.
Chief Executive and co-founder Aaron Ginn said the Series A will help Hydra Host broaden its data center presence, develop its platform and strengthen global operations as customers seek easier access to GPUs across markets. Kindred Ventures founder and managing partner Steve Jang said the company’s combined operating system and offtake network could offer an asset-light model for connecting an expanding supply of AI data centers with demand for compute-intensive workloads. With the funding in place, Hydra Host is aiming to deepen its role in the AI infrastructure value chain by helping operators translate power, facilities, hardware and capital into revenue-generating AI capacity.