# GPU Cloud (gpucloud.ca)

Rents hourly NVIDIA GPU servers hosted in Canada, with SSH access and preconfigured images for training and inference.

- Atlas card: https://inite.ai/en/atlas/gpucloud.ca
- Site: https://gpucloud.ca
- For: AI developers and enterprises needing Canada-hosted GPUs for model training and inference
- Categories: llmops
- Use not measured/100 · alive 100/100

## Used for

- fine-tune a language model on an hourly-billed GPU
- serve open models through an OpenAI-compatible API
- generate images with ComfyUI on a rented GPU

## Price

- from 0.63 CAD/usage, no free plan (read 2026-10-21 on https://gpucloud.ca/fr/gpu/tarifs)

## For developers

- API docs: https://gpucloud.ca/fr/developers
- MCP server: ca.gpucloud/gpu-cloud (namespace; open on 2026-10-21, 62 tools)
- Read 2026-10-21

## MCP server, measured daily

Server: https://gpucloud.ca/api/mcp, checked every day since 2026-10-21 (2 checks).

- Answered: measuring, 2 daily checks so far
- Median first answer: 323 ms
- Sign-in: none
- Context its tool definitions take: about 11533 tokens (62 tools)
- Tool definitions: A, 95/100
- Failing "params_described": 19 of 62 tools
- Warning "open_writes": gpucloud_update_profile, gpucloud_create_ticket, gpucloud_set_notices, gpucloud_create_server, gpucloud_delete_server
- Warning "destructive_unmarked": gpucloud_cancel_reservation
- Method: https://inite.ai/en/atlas/mcp/report

## Your data

- Certifications stated: none
- Read 2026-10-21

---

Source: INITE Atlas, https://inite.ai/en/atlas/gpucloud.ca. JSON: https://inite.ai/api/atlas/gpucloud.ca. MCP: https://inite.ai/api/mcp (atlas_lookup).