Use cases
AI compute for fine-tuning

Fine-tuning GPU Workspaces

Provision workspaces for adapter training, supervised fine-tuning, evaluation, and model packaging.

Ready template

Recommended workspace

Template

PyTorch

GPU

H100

Access

Browser + SSH

Billing

Per-hour

porfal create --use-case fine-tuningworkspace ready: browser + ssh

Workloads

LoRA

QLoRA

SFT runs

Evaluation jobs

Recommended stack

PyTorch

JupyterLab

Ubuntu CUDA

vLLM

GPU fit

H100

H200

RTX 6000 Ada

Why Porfal

Dedicated GPU compute without infrastructure work.

Pick a GPU, launch the template, connect from browser or SSH, and stop the workspace when the run is done.

Use high-VRAM GPUs only when fine-tuning runs need them.

Keep training artifacts isolated per workspace.

Move from fine-tuning to inference testing quickly.

Workflow

A short path from GPU to running workload.

1

Choose high VRAM

2

Prepare dataset

3

Run fine-tune

4

Evaluate model