What you get
- Four GPU types, 1 to 4 GPUs per machine.
t4,l4, anda100machines take 1, 2, or 4 GPUs, and anh100machine takes one. Each GPU brings its own vCPUs and memory. - A machine that is ready to use. CUDA, PyTorch, the Hugging Face
hfCLI, anduvare installed, andnvidia-smishows your GPU as soon as the machine runs. - A 200 GiB disk for models and data. Your home,
/data/home, is on it, so a model that you download once is still there after a pause or a stop. - Pause and resume with the model loaded. A pause keeps the machine’s processes and the GPU’s
memory, and a paused machine does not pay for its GPU. An
h100machine cannot be paused. - An HTTPS URL for a model server. Publish a port to call your model from anywhere, or keep the URL private behind a token.
- Available on every paid plan, with up to running GPUs per workspace.
Start a GPU machine
Name the GPU type when you create the machine. InstaBox picks the template and the size for that type.--gpu l4:2.
Workloads
Each guide starts from an empty machine and ends with a working result on onel4 GPU, with the
output of a real run.
Serve an open model
Serve an open LLM with vLLM and call it with any OpenAI client.
Train a model
Train as a background job, pause without losing the run, and continue from a checkpoint.
Generate images
Turn prompts into pictures with an open image model.
Transcribe audio and describe video
Run Whisper and a vision model over your own files.
Run batch jobs
Run a model over a large file of prompts, or split a sweep across many GPUs.
Pick a GPU
The GPU’s own memory decides which models fit.
When a model does not fit one GPU, give the machine more GPUs and split the model across them.
Cost
GPUs are billed per GPU, by the second, only while the machine runs. For example, one costs an hour all in, with its vCPUs and memory. A paused or stopped machine holds no GPU and pays only for its disk. See Pricing.A machine pauses after idle minutes by default. Its processes and its loaded
model are kept, and a request to a public port wakes it. To keep a machine running, set a longer
idle time or turn the idle pause off. See Auto-pause.
Limits
- When every GPU of a type is in use,
createexits 75 withno_capacity. Try again later, or pick another GPU type. - Docker does not run in a GPU machine. Install a model server with
piporuv, as the guides do. - A GPU machine cannot be forked. To copy one, save a snapshot and create machines from it.
- What works differently lists the other differences from a machine without a GPU.
GPU machines
GPU types and counts, the template, and model weights in full.
Pricing
The price of each GPU type.