HeatLoop
et
For compute buyers

GPU capacity from infrastructure that sells its heat instead of dumping it.

HeatLoop installs liquid-cooled NVIDIA accelerators inside Estonian buildings that buy the recovered heat. Same silicon, EU jurisdiction, and a cost structure with one fewer expense in it — because the cooling is somebody else's heating.

Talk to us about capacity

How capacity is allocated

Capacity comes online as the first sites are commissioned, and it is allocated ahead of build.

Capacity is allocated against committed demand rather than sold from an anonymous pool. Tell us the workload — size, duration, and whether it can checkpoint — and we will come back with a configuration, a lead time and a rate, or tell you plainly that we are not the right fit. Committed demand is what sets the configuration and the deployment order.

What makes this different

The heat is a product, not a cost

Conventional facilities spend capital and power on rejecting heat. We sell ours to the building the rack sits in. That removes a cost line rather than passing it through to you.

Inside the EU

Hardware, operations and company are Estonian. For teams with data-residency requirements, that is a shorter compliance conversation than a non-EU region flag.

You talk to the operator

Capacity questions, scheduling and incidents go to the people who run the fleet, not to a ticket queue. We deal direct, and intend to keep it that way.

Hardware

Rack configurations are specified around NVIDIA workstation-class RTX PRO accelerators and data-centre-class DGX systems, sourced new and certified-refurbished through authorised channel partners.

Each rack draws about 166 kW fully populated and is liquid-cooled end to end. Configuration is set per deployment against the workloads it is being built for, rather than fixed to a standard catalogue product.

Workloads this suits
Model training and fine-tuning
Multi-day runs on dedicated hardware with persistent storage, reserved for the length of the job.
Inference
Steady-state serving on committed capacity, with latency inside the EU for European users.
Rendering and simulation
Frame and batch queues from studios and engineering teams — naturally parallel, naturally schedulable.
Checkpointable batch
Jobs that can pause and resume run on interruptible capacity at a lower rate, because we can schedule them into the cheapest hours of the day.

Two ways to buy

Rates are quoted against the configuration and the power contract behind it, not from a public price list. Tell us the workload and we will put a real number in front of you.

Committed capacity

Reserved GPUs for a fixed term. Predictable performance and predictable cost, for training runs and production inference.

  • Dedicated hardware
  • Fixed monthly rate
  • Persistent storage
  • Term from one month

Interruptible capacity

Lower rate in exchange for jobs that checkpoint and resume. Suited to rendering queues and batch training where wall-clock time is flexible.

  • Reduced rate
  • Advance notice before pause
  • Checkpoint-and-resume required
  • Billed per GPU-hour used

Tell us what you need to run

Describe the workload and we will tell you honestly whether we suit it. No obligation, and what you ask for genuinely shapes how the capacity is configured.

  • What you would run — training, inference, rendering or batch
  • Roughly how many GPUs, and for how long
  • Whether your jobs can checkpoint and resume
  • Any data-residency or compliance requirements
  • When you would need capacity
Email us about capacity

info@heatloop.com