Train and run models on a cloud GPU.

Preinstalled — the machine starts from the official PyTorch image

Notebooks, a terminal and a file browser, opened from your dashboard

pip install transformers to load and run open language models

pip install diffusers for image-generation models such as Stable Diffusion

Upload your code and run it from a terminal or over SSH

Fine-tune models with your own data, e.g. LoRA with pip install peft

You are root in the container — install the libraries you need

Start jobs with nohup so they keep running after you disconnect