SF Compute exists to give ambitious teams serious compute without overbuying, underutilizing, or taking on unnecessary risk.
We build data centers, the GPU clusters inside them, & the cloud on top. Get VMs, Bare Metal, managed Slurm & Kubernetes from the only cloud designed to let you sublease your reserved capacity when plans change.
Reserve the compute you need and resell what you don't, so idle GPUs become revenue instead of sunk cost. Our scheduler allows tenants to sell back their idle compute to others.
We support contracts ranging from one node for one hour to thousands of nodes for multiple years. Customers can access compute via our API, CLI, or by working directly with our team for larger deployments. We offer virtual machines, bare metal, and managed slurm.
For long-term commitments, reserve a custom H100, B300, or GB300 cluster with us. View our available clusters and reach out to secure a contract.
No dashboards to learn. Install sf, log in, and buy or resell GPU capacity right from the terminal. Plain commands you can script and automate.
We're a small, dense team in San Francisco building the order book and running the GPU clusters behind it. If you ship fast and enjoy solving hard systems problems, we'd love to talk.
