Installation
LynxDB ships as one statically linked binary (CGO_ENABLED=0) with no
runtime dependencies.
Pre-built binary
- Linux
- macOS
- Windows
curl -fsSL https://lynxdb.org/install.sh | sh
The script detects the architecture (amd64, arm64, armv7) and the C library
(glibc or musl), downloads the matching release archive, checks it against
the release's SHA-256 checksums when sha256sum or shasum is available,
and installs to /usr/local/bin, or to ~/.local/bin when /usr/local/bin
is not writable.
# Install script (Intel and Apple Silicon)
curl -fsSL https://lynxdb.org/install.sh | sh
# Or Homebrew
brew install lynxbase/tap/lynxdb
Download the windows-amd64 zip archive from
GitHub Releases, or build it
with Go:
go install github.com/lynxbase/lynxdb/cmd/lynxdb@latest
The install script also runs under Git Bash and MSYS2.
Install script options
Pass options after sh -s --, or set the matching environment variable:
| Option | Variable | Default | Description |
|---|---|---|---|
--version <v> | LYNXDB_VERSION | latest | Install a specific version, such as v0.5.0 |
--channel <c> | LYNXDB_CHANNEL | stable | Release channel: stable or nightly |
--allow-prerelease | LYNXDB_ALLOW_PRERELEASE | unset | Required for nightly and prerelease versions |
--dir <path> | LYNXDB_INSTALL_DIR | auto | Installation directory |
--no-modify-path | LYNXDB_NO_MODIFY_PATH | unset | Do not add the directory to PATH |
--force | LYNXDB_FORCE | unset | Skip confirmation prompts |
LYNXDB_BASE_URL | https://dl.lynxdb.org | Download from another base URL (GitHub Releases is the fallback) |
# A specific version
curl -fsSL https://lynxdb.org/install.sh | sh -s -- --version v0.5.0
# A custom directory
curl -fsSL https://lynxdb.org/install.sh | LYNXDB_INSTALL_DIR=~/bin sh
Nightly builds
Nightly builds are prerelease builds of main for testing. They need an
explicit opt-in.
# Install script
curl -fsSL https://lynxdb.org/install.sh | sh -s -- --channel nightly --allow-prerelease
# Homebrew
brew install lynxbase/tap/lynxdb-nightly
To return to the stable channel, run the plain install script again, or swap the Homebrew formula:
brew uninstall lynxdb-nightly
brew install lynxbase/tap/lynxdb
Docker
Images are published to ghcr.io/lynxbase/lynxdb for linux/amd64 and
linux/arm64. The image runs lynxdb server by default. Inside a container,
listen on 0.0.0.0 and keep the data directory on a volume:
docker run -d --name lynxdb \
-p 3100:3100 \
-v lynxdb-data:/data \
ghcr.io/lynxbase/lynxdb server --addr 0.0.0.0:3100 --data-dir /data
Run a one-off query over standard input:
echo '{"level":"error","msg":"test"}' | \
docker run -i --rm ghcr.io/lynxbase/lynxdb query 'source=main | stats count() by level'
The nightly tag follows the nightly builds:
docker pull ghcr.io/lynxbase/lynxdb:nightly
Docker Compose
services:
lynxdb:
image: ghcr.io/lynxbase/lynxdb:latest
command: server --addr 0.0.0.0:3100 --data-dir /data
ports:
- "3100:3100"
volumes:
- lynxdb-data:/data
volumes:
lynxdb-data:
See Docker deployment for more examples.
From source
Requires Go 1.26 or later:
# go install
go install github.com/lynxbase/lynxdb/cmd/lynxdb@latest
# Or clone and build
git clone https://github.com/lynxbase/lynxdb.git
cd lynxdb
go build -o lynxdb ./cmd/lynxdb/
Verify the installation
lynxdb version
Version v0.5.0
Commit abc1234
Built 2026-03-01T10:00:00Z
Go go1.26.3 linux/amd64
lynxdb version --short prints only the version; --json prints the same
information as JSON.
Run as a service
The binary does not install a service unit. To run the server under systemd, see Server Mode and Single-node deployment.
Shell completion
# Bash
lynxdb completion bash >> ~/.bashrc
# Zsh
lynxdb completion zsh >> ~/.zshrc
# Fish
lynxdb completion fish > ~/.config/fish/completions/lynxdb.fish
PowerShell is supported too: lynxdb completion powershell.
Supported platforms
| OS | Architecture | Archives |
|---|---|---|
| Linux | x86_64 (amd64) | glibc and musl |
| Linux | arm64 | glibc and musl |
| Linux | armv7 | glibc |
| macOS | Intel (amd64) | |
| macOS | Apple Silicon (arm64) | |
| Windows | x86_64 (amd64) | zip |
Next steps
- Quick Start: run a first query
- Pipe Mode: query files without a server
- Server Mode: start a persistent server