Installation

Requirements

• Linux x86_64 or Windows 11, AMD gfx1151 (Ryzen AI Max 300 and 400 series — Strix Halo and Gorgon Halo). WSL2 works; ROCm reaches the GPU there through /dev/dxg, so set it up with AMD's WSL guide first
• ROCm 7.x.
• Python 3.10 – 3.14, and ROCm builds of torch and torchvision — see below.

Choose an installation method

System	Installation method
Linux	Install script or manual installation
WSL2	Install script or manual installation inside WSL2
Windows 11 (native)	Manual installation only

Linux / WSL2: install script

Run this command in your Linux or WSL2 shell. The install script does not support native Windows.

curl -fsSL https://snowllm.dev/install.sh | sh

• Checks the machine before downloading anything: GPU, driver access, Python, disk, network
• Installs into ~/.local/share/snowllm, links snowllm into ~/.local/bin (XDG or SNOWLLM_* overrides can change these paths)
• Re-run to upgrade. Remove with curl -fsSL https://snowllm.dev/install.sh | sh -s -- --uninstall.

Linux / WSL2: manual installation

Alternatively, install manually. Torch and torchvision must come from AMD's index:

# https://rocm.docs.amd.com/projects/ai-ecosystem/en/latest/frameworks/pytorch/install.html
pip install --index-url https://repo.amd.com/rocm/whl-multi-arch/ \
    "torch[device-gfx1151]==2.12.0+rocm7.14.0" \
    "torchvision==0.27.0+rocm7.14.0"
pip install snowllm snowllm-kernels

Windows: manual installation

Native Windows supports manual installation only. Run the following commands in PowerShell or Command Prompt using your Python environment. Torch and torchvision must come from AMD's index:

python -m pip install --index-url https://repo.amd.com/rocm/whl-multi-arch/ "torch[device-gfx1151]==2.12.0+rocm7.14.0" "torchvision==0.27.0+rocm7.14.0"
python -m pip install snowllm snowllm-kernels

If you are using WSL2 on a Windows machine, follow the Linux / WSL2 instructions above inside WSL2 instead.

Next step

Download your first model and start chatting.