Tailor big models on your timetable, not the budget’s
Launch the stack you know and keep costs under control. Pause and resume without losing your setup.
Why teams choose Compute for training
Ready images
Ubuntu, PyTorch, and Jupyter Notebook integration to get moving fast.
Custom templates
Lock in your environment and reuse it for future runs.
Stop/Start (limited-time no extra fees)
Pause a long run, then continue later.
All-inclusive pricing
No egress charges. Per-second billing with prepaid credits and auto top-up.
How it works
Pick a 4090 or 5090 tier.
Launch with PyTorch or Ubuntu, open Jupyter if you like.
Move your data into the instance using your own scripts or notebook.
Train, monitor, and save as a custom template for next time.
Practical patterns
Quick LoRA experiments on a single 4090
Heavier fine-tunes on a 5090
Notebooks for exploration, then headless training
Pricing at a glance
On-demand GPUs, billed per second via prepaid credits
All-inclusive: compute, storage, and data transfer included
Welcome bonus: up to €250 on first purchase
RTX 5090
RTX 4090
GPUs are on-demand today. Spot capacity is coming soon.
Do you have any questions?
Do you support multi-GPU?
Yes. Launch larger tiers when you need more headroom.
Can I bring my own container?
Yes. Use your image and scripts within the instance.
Can I cap spend?
Yes. Prepaid credits with auto top-up and your own limits.
Can I bring my own Docker image?
Yes. Push to your registry and point the run at it.
