Download

Pick how you install.

LocalAI is one binary or one container. Choose what fits your machine.

Install methods

Script for Linux, and for macOS on Apple Silicon. It fetches the latest release binary.
Container: Docker or Podman, with the localai/localai image.
macOS app as a DMG. See the macOS guide.
Kubernetes with a Helm chart or a manifest.
Build from source as described in the build guide.

shell

Script

$ curl -sSL https://localai.io/install.sh | sh
$ local-ai run qwen3-4b
# the API and the web UI are on http://localhost:8080

Docker

$ docker run -p 8080:8080 --name local-ai -ti localai/localai:latest
# Podman works the same way

Kubernetes

$ helm repo add go-skynet https://go-skynet.github.io/helm-charts/
$ helm install local-ai go-skynet/local-ai
# or: kubectl apply -f https://localai.io/install/kubernetes.yaml

More: containers and GPU images, the macOS app, Linux, Helm chart, build from source

The script, the macOS app and the latest container image give version 4.12.1. Office, Sandbox, Team and add-ons need a development build or a newer release.

Hardware

Hardware	What to know
CPU	Works without a GPU.
NVIDIA	CUDA images and backends.
AMD	ROCm images and backends.
Intel GPU	SYCL and oneAPI.
Vulkan	Documented in the GPU guide.
Apple silicon	Metal, with the macOS app or binary.
NVIDIA Jetson	Documented in the container guide.

Read the GPU guide