Local AI › How to install local AI on Linux, step by step

How to install local AI on Linux, step by step

From easiest to most technical. Each step comes from the tool's official instructions, linked in the text.

Checked on 13 Oct 2026. Versions change every week: check the official site before installing.

1. LM Studio

Download the AppImage (or the .deb) from lmstudio.ai/download, make it executable with chmod +x LM-Studio-*.AppImage and run it with ./LM-Studio-*.AppImage (the exact file name may vary). For servers there is llmster, a daemon with no graphical interface (same download page).

More about LM Studio

2. Ollama

Run curl -fsSL https://ollama.com/install.sh | sh (official instructions).

First use, on any system: ollama run gemma4 or ollama run qwen3.5:4b. ollama list shows the models you have installed. The local API listens on http://localhost:11434 (README). Models marked «-cloud» in the Ollama library do not run on your computer.

More about Ollama

3. Jan

Download the installer from jan.ai or from the links in the README (Windows jan.exe, macOS jan.dmg, Linux jan.deb / jan.AppImage / Arm64), open «Hub», download a model and start chatting.

More about Jan

4. Open WebUI

With Python: pip install open-webui and then open-webui serve, and open http://localhost:8080. With Docker and Ollama on the same machine, the README gives: docker run -d -p 3000:8080 --add-host=host.docker.internal:host-gateway -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:main and then open http://localhost:3000. The same README has variants for GPUs and for Ollama on another server.

More about Open WebUI

5. llama.cpp

Official installer (current README): macOS/Linux curl -LsSf https://llama.app/install.sh | sh; Windows (PowerShell) irm https://llama.app/install.ps1 | iex. There are also binaries under Releases and a winget package. The README uses the unified llama command: llama cli -hf ggml-org/Qwen3.5-0.8B-GGUF (chat in the terminal) or llama serve -hf ggml-org/Qwen3.5-0.8B-GGUF (server with a web interface); -hf downloads the model from Hugging Face.

More about llama.cpp

Images

To generate images, see Generate AI images locally with ComfyUI.