Installation

Download

Download the right file for your system from the GitHub Releases page:

Your Computer	File Name
Windows (most PCs)	`SwarmLLM-Setup.exe` (installer — auto-detects GPU)
Windows (raw binary, GPU)	`swarmllm-windows-x86_64-gpu.zip`
Windows (raw binary, CPU)	`swarmllm-windows-x86_64-cpu.zip`
Mac (M1/M2/M3/M4)	`swarmllm-macos-aarch64.tar.gz` (compile-validated)
Mac (older Intel)	Best-effort — build from source
Linux (most distros)	`swarmllm-linux-x86_64.tar.gz`
Linux (NVIDIA GPU)	`swarmllm-linux-x86_64-cuda.tar.gz`

Not sure which Mac? Apple menu > "About This Mac." If it says "Apple M1" (or M2/M3/etc.), pick Apple Silicon. If it says "Intel," pick Intel.

Recommended — installer: double-click SwarmLLM-Setup.exe. It detects your GPU (NVIDIA / AMD / Intel) and installs the matching binary. If SmartScreen warns you, click More info > Run anyway.

Raw binary alternative: download swarmllm-windows-x86_64-gpu.zip (Vulkan + CUDA static) or swarmllm-windows-x86_64-cpu.zip (CPU-only fallback), extract, and run swarmllm.exe.

From PowerShell on a raw binary:

cd Downloads\swarmllm-windows-x86_64-gpu
.\swarmllm.exe run

macOS

cd ~/Downloads
tar xzf swarmllm-macos-aarch64.tar.gz
cd swarmllm-macos-aarch64
chmod +x swarmllm
./swarmllm run

Note: macOS aarch64 binaries are compile-validated and exercised in CI (test + clippy on macos-15); integration tests stay Linux-only for now. Intel Mac users should build from source. If macOS blocks the binary on first launch: System Settings > Privacy & Security > click Open Anyway next to SwarmLLM.

Linux

cd ~/Downloads
tar xzf swarmllm-linux-x86_64.tar.gz
cd swarmllm-linux-x86_64
chmod +x swarmllm
./swarmllm run

Docker

The fastest way to get running on any Linux server:

# 1. Get the compose file and example env
curl -LO https://raw.githubusercontent.com/enapt/SwarmLLM/main/docker-compose.yml
curl -LO https://raw.githubusercontent.com/enapt/SwarmLLM/main/.env.example

# 2. Configure (add API keys, change ports, etc.)
cp .env.example .env
nano .env

# 3. Start
docker compose up -d

For NVIDIA GPU support (requires NVIDIA Container Toolkit):

docker compose --profile gpu up -d

Pre-built images on GHCR:

Image	Description
`ghcr.io/enapt/swarmllm:latest`	CPU-only
`ghcr.io/enapt/swarmllm:latest-cuda`	NVIDIA GPU (CUDA 12.4)
`ghcr.io/enapt/swarmllm:0.1.0`	Pinned version (CPU)
`ghcr.io/enapt/swarmllm:0.1.0-cuda`	Pinned version (GPU)

Data is persisted in Docker volumes. Model shards are stored in the swarmllm-models volume (or bind-mount a host directory via SWARMLLM_MODELS_DIR in .env).

View logs with docker compose logs -f. The API key is printed on first startup.

Cargo Install

Requires Rust 1.80+:

cargo install --git https://github.com/enapt/SwarmLLM.git --tag v0.1.0
swarmllm run

Building from Source

git clone https://github.com/enapt/SwarmLLM.git
cd SwarmLLM
cargo build --release
./target/release/swarmllm run

For CUDA GPU support:

cargo build --release --features candle-cuda

For Apple Silicon: the default build runs on CPU. A Metal-accelerated build is on the roadmap but not yet implemented (no metal Cargo feature exists yet); until then, use the default cargo build --release.

Open the Dashboard

Once running, open http://localhost:8800 in your browser. The setup wizard will walk you through initial configuration.

SwarmLLM Documentation