Install Spice.ai OSS
Installation options for Spice.ai OSS
Prerequisites:
- macOS 14+ (Apple Silicon), or Linux (x86_64 or aarch64) with glibc 2.39+, such as Ubuntu 24.04+ or Debian 13+. See System Requirements.
- Windows: use WSL 2 with Ubuntu 24.04 (
wsl --install -d Ubuntu-24.04), then follow the Linux steps. The Spice runtime does not run natively on Windows. curl, used by the install script. Minimal Linux images and containers often don't include it. On Ubuntu or Debian, install it withsudo apt-get install -y curl.
For deployment options, such as to Kubernetes, see Deployment.
- macOS, Linux, and WSL
- Windows
Install Script​
curl https://install.spiceai.org | /bin/bash
Homebrew​
brew install spiceai/spiceai/spice
To run the Spice runtime locally on Windows, use WSL 2 and follow the macOS, Linux, and WSL tab from the WSL terminal. The PowerShell script below installs only the native spice CLI, for example to use with Spice Cloud.
PowerShell Install Script​
iex ((New-Object System.Net.WebClient).DownloadString("https://install.spiceai.org/Install.ps1"))
Direct Download​
Binaries for Linux, Windows, and macOS are available for download from GitHub at github.com/spiceai/spiceai/releases.
Verify the installation:
After installing, verify Spice is installed correctly:
spice version
Expected output (the runtime is downloaded on the first spice run):
CLI version: v2.x.x
Runtime version: not installed
If the command is not found, the current terminal hasn't picked up the updated PATH yet. Open a new terminal, or add the Spice binary directory to the PATH of the current shell:
export PATH="$HOME/.spice/bin:$PATH"
On Linux, the install script can add the PATH entry to ~/.profile, which some terminals don't read. If spice still isn't found in new terminals, add the line above to ~/.bashrc (or ~/.zshrc for zsh).
What's Next?​
After installing, follow the Getting Started guide to initialize a Spice app, connect to a dataset, and run your first query in under 5 minutes.
Building from Source​
To build Spice from source, including CUDA and Metal hardware acceleration options, see CONTRIBUTING.md in the Spice.ai GitHub repository.
