Dedicated GPUs

Train or run your model in Canada.

Canadian GPUs run by Northern Inference and dedicated to your model. Tell us your model and usage; we reply by email with a custom quote and a date.

Request a quote How it works

What it is

  • GPUs we run in Canada.
  • Dedicated to training or running your model.
  • You choose the model. We plan the setup with you.

Who it is for

  • Teams training or fine-tuning a model.
  • Teams that need capacity no one else shares.
  • Work that needs Canadian data residency.
  • Models that need dedicated hardware to run.

Custom quote

  • Tell us your model, expected usage and timing.
  • We reply by email with a custom quote and a date.
  • Your request is free and commits you to nothing.

How long it takes

  • Usually online within 4 to 8 weeks.
  • Your quote confirms the proposed date.

After you send a request

  • We review what you need to train or run your model.
  • A request is free and commits you to nothing.
  • Nothing is set up until you accept.

How it works

  1. Name your modelTell us its name or share a link, including a model of your own.
  2. Tell us what you needTell us about training or running your model, expected usage and timing.
  3. Get a quote and a dateWe reply by email. You decide.

Tell us about your model

Your model trains or runs on GPUs we run in Canada.

This request is free and commits you to nothing. We reply by email with a quote and a date.

Prefer email? Write to hello at northerninference dot ca.

We use these details only to reply to your request. See our Privacy Policy.

↑