This page explains each setting on the deploy page and its CLI flag, and what happens after you launch.
On the Instances page, click Deploy instance.
/workspace. It's erased when the instance stops or restarts, and it's billed only while the instance runs./workspace. A volume keeps its data when instances stop or are deleted, and several instances can share it. Volumes can't be added or removed after you deploy, and an instance that mounts one runs in the volume's location.HTTP ports in the container that get their own public HTTPS address, for a notebook, an API or a web app. Anyone with the address can reach them. Only HTTP traffic is carried, and you can change ports later without a restart. See Expose ports.
/run/featherless/secrets, not as environment variables. They can't be changed after you deploy. See Secret variables.featherless-cloud instance create \
--name my-instance \
--template featherless/rocm-pytorch \
--gpu-count 1 \
--disk 100 \
--volume my-data:/workspace \
--port 8888 \
--env HF_HOME=/workspace/hf
For CPU compute, use --compute-type cpu --cpu-cores 8. To choose a GPU model, add --gpu-type <model>; featherless-cloud pricing list shows the models on offer.
See instance create for every option.
Before the container starts, the instance waits for a GPU and downloads its image. The first start of an image can take a few minutes; later starts are faster. Billing is per second and starts when the container starts, not while you wait.