From a single GPU to a whole cluster
For teams that run their own models: launch a virtual machine with root access, train without managing machines, or reserve a cluster with our engineers.
Which option fits your workload
Reserve capacity for the long run, or launch a GPU when you need one.
Reserved GPUs and clusters
Current capacity
Nothing listed fits your plan yet. Book a call and Benedikt will look for it, or add a backup GPU.
GPU virtual machines
Launch a virtual machine with root access in our European data centres and install what you need. You pay per second while it runs.
GPU prices and capacity
Size a model
See the memory a Hugging Face model needs and how many requests it serves.
Serverless training
Train models without managing machines: package your training code in a container, choose a GPU and submit the job.
You
Package your code
Put your training code and its dependencies in a plain Docker container.
You
Choose a GPU and submit
Pick the GPU your model needs, check its price, then submit the workload.
We
We run it on that GPU
The job runs on the hardware you chose. When it finishes, you inspect its output.
# 1 Package your code
$ docker build -t your-team/train .
$ docker push your-team/train# 2 Choose a GPU and submit
$ lyceum docker run your-team/train \
-m gpu.h100 -c "python train.py"# 3 We run it on that GPU
# The output streams here until
# the run ends. Add -d to detach.