Compute

Compute Overview

Deploy GPU and CPU instances directly from your DEVUP AI dashboard — full root access via SSH and Jupyter, per-second billing, and persistent storage options.

DEVUP AI Compute lets you launch cloud instances on demand. Choose between GPU instances for machine learning, training, and GPU-accelerated workloads, or CPU instances for general-purpose compute, scripting, and lightweight tasks.

GPU vs CPU — when to use which

Dimension	GPU Instances	CPU Instances
Best for	ML training, fine-tuning, inference, rendering, GPU-accelerated compute	Data processing, scripting, web scraping, CI/CD, general-purpose workloads
Hardware	NVIDIA GPUs (RTX, A-series, H-series) with dedicated VRAM	Multi-core vCPUs with configurable RAM
Storage	Container Disk + Volume Disk (ephemeral) or Network Volumes (persistent)	Container Disk (ephemeral) or Network Volumes (persistent)
Billing	Per-second while running, reduced idle rate while paused	Per-second while running, reduced idle rate while paused
Access	SSH + JupyterLab	SSH + JupyterLab

Resource sizing guide

When selecting an instance size for Large Language Models (LLMs), a general rule of thumb is ~2GB of VRAM per billion parameters for unquantized models. For example:

7B models: ~14GB+ VRAM
13B models: ~26GB+ VRAM
70B models: ~140GB+ VRAM (or ~35GB+ when using 4-bit quantization)

Use tools like the Hugging Face Model Memory Calculator to estimate exact requirements.

Workload	Recommended Tier
LLM Inference	Mid-range to High-end GPU
Training / Fine-Tuning	High-end GPU (multi-GPU for large models)
Image Generation & 3D Rendering	Entry to Mid-range GPU
Computer Vision	Mid-range GPU
Data Processing & Scripting	CPU Instances

Availability depends on your selected data center. See the GPU Deploy and CPU Deploy pages for current options.

How it works

Choose a tier — select a GPU model or CPU flavor from the live catalog in your dashboard. Availability and pricing are shown in real time.
Configure your instance — pick a container template, set your storage allocation, and optionally attach a persistent Network Volume.
Deploy — your instance is provisioned automatically. Once running, connect via SSH or JupyterLab.
Manage — pause to save costs (a reduced idle rate applies), resume when ready, or terminate to stop billing entirely.

Prerequisites

A DEVUP AI account with a funded balance (top up from Billing).
For SSH access: an SSH public key saved in Account Settings → Security.

Next steps

Deploy a GPU Instance

Launch a GPU-accelerated instance for ML and compute workloads.

Deploy a CPU Instance

Spin up a general-purpose CPU instance for lightweight workloads.

Compute Overview

Deploy GPU and CPU instances directly from your DEVUP AI dashboard — full root access via SSH and Jupyter, per-second billing, and persistent storage options.

GPU vs CPU — when to use which

Dimension	GPU Instances	CPU Instances
Best for	ML training, fine-tuning, inference, rendering, GPU-accelerated compute	Data processing, scripting, web scraping, CI/CD, general-purpose workloads
Hardware	NVIDIA GPUs (RTX, A-series, H-series) with dedicated VRAM	Multi-core vCPUs with configurable RAM
Storage	Container Disk + Volume Disk (ephemeral) or Network Volumes (persistent)	Container Disk (ephemeral) or Network Volumes (persistent)
Billing	Per-second while running, reduced idle rate while paused	Per-second while running, reduced idle rate while paused
Access	SSH + JupyterLab	SSH + JupyterLab

Resource sizing guide

When selecting an instance size for Large Language Models (LLMs), a general rule of thumb is ~2GB of VRAM per billion parameters for unquantized models. For example:

7B models: ~14GB+ VRAM

13B models: ~26GB+ VRAM

70B models: ~140GB+ VRAM (or ~35GB+ when using 4-bit quantization)

Workload	Recommended Tier
LLM Inference	Mid-range to High-end GPU
Training / Fine-Tuning	High-end GPU (multi-GPU for large models)
Image Generation & 3D Rendering	Entry to Mid-range GPU
Computer Vision	Mid-range GPU
Data Processing & Scripting	CPU Instances

Availability depends on your selected data center. See the GPU Deploy and CPU Deploy pages for current options.

How it works

Choose a tier — select a GPU model or CPU flavor from the live catalog in your dashboard. Availability and pricing are shown in real time.

Configure your instance — pick a container template, set your storage allocation, and optionally attach a persistent Network Volume.

Deploy — your instance is provisioned automatically. Once running, connect via SSH or JupyterLab.

Manage — pause to save costs (a reduced idle rate applies), resume when ready, or terminate to stop billing entirely.