NVIDIA Brev CLI

NVIDIA Brev provides streamlined access to NVIDIA GPU instances on popular cloud platforms, automatic environment setup, and flexible deployment options, enabling developers to start experimenting instantly.

Install the cli

MacOS

Assumes Homebrew (or Workbrew equivalent) are installed.

brew install brevdev/homebrew-brev/brev

Linux

curl -fsSL https://raw.githubusercontent.com/brevdev/brev-cli/main/bin/install-latest.sh | bash

Installs to ~/.local/bin. If it's not on your PATH, add this to your shell profile:

export PATH="$HOME/.local/bin:$PATH"

To install somewhere else, set BREV_INSTALL_DIR for the installer:

curl -fsSL https://raw.githubusercontent.com/brevdev/brev-cli/main/bin/install-latest.sh | BREV_INSTALL_DIR="$HOME/bin" bash

Windows

Using Brev With Windows Subsystem for Linux (WSL)

Brev is supported on windows currently through the Windows Subsystem for Linux (WSL). This guide will walk you through the steps to get Brev up and running on your Windows machine.

Prerequisites

• WSL installed and configured
• Virtualization enabled in your BIOS
• Ubuntu >=22.04 installed from the Microsoft Store

Once WSL is set up, install Brev with:

curl -fsSL https://raw.githubusercontent.com/brevdev/brev-cli/main/bin/install-latest.sh | bash

Installs to ~/.local/bin. If it's not on your PATH, add this to your shell profile:

export PATH="$HOME/.local/bin:$PATH"

From conda-forge

To globally install brev from conda-forge in an isolated environment with Pixi, run

pixi global install brev

Get Started

Log in to your Brev account:

brev login

Create a new GPU instance:

brev create awesome-gpu-name

See the instance:

brev ls

Docs

https://docs.nvidia.com/brev/latest/

---

AI Agent Integration

Brev CLI includes a skill for AI coding agents (like Claude Code) that enables natural language GPU instance management.

# Install via CLI
brev agent-skill

# Or via standalone installer
curl -fsSL https://raw.githubusercontent.com/brevdev/brev-cli/main/scripts/install-agent-skill.sh | bash

Once installed, you can say things like "create an A100 instance for ML training" or "search for GPUs with 40GB VRAM" in your AI coding agent.

Contributing

We welcome PRs! Checkout Contributing.md for more.