Ollama & LM Studio Beginner’s Guide
Pick your computer and an app, then tick off each step. Every step comes from the app’s own documentation.
- ollama.com/downloadDownload for Windows OllamaSetup.exe
Run the installer
Open OllamaSetup.exe. It installs into your home folder and doesn’t need administrator rights.
Ollama for WindowsRun your first model
Open PowerShell and run the model from Ollama’s quickstart. The first run downloads it, then a chat starts.
Ollama quickstartPowerShellPS> ollama run gemma4:e2bpulling manifest …- PowerShell>>> Write a haiku about autumn…>>> /bye
Frequently asked questions
Should I use Ollama or LM Studio?
Both run AI models on your own computer. Ollama is used by typing commands in a terminal (PowerShell on Windows, Terminal on a Mac), and also comes with an app. LM Studio is a desktop app: you find models in its Discover tab and chat with them in its Chat tab, without typing commands.
What do I need to run Ollama?
On Windows, Ollama’s documentation lists Windows 10 22H2 or newer (Home or Pro), at least 4 GB of disk space for the app, and up-to-date graphics drivers: NVIDIA 551.61 or newer, or for AMD a ROCm v7 / HIP7 or Vulkan-capable Radeon driver. On a Mac it needs macOS Sonoma (14) or newer; Apple M-series Macs use the CPU and GPU, Intel Macs the CPU only.
What do I need to run LM Studio?
LM Studio’s documentation lists, on Windows, an x64 processor with AVX2 or an ARM Snapdragon X Elite, with 16 GB of RAM and 4 GB of dedicated graphics memory recommended. On a Mac it needs Apple silicon (M1 to M4) and macOS 14.0 or newer, with 16 GB of RAM recommended. Intel Macs aren’t supported.
How much disk space do models take?
Ollama’s documentation says models can be tens to hundreds of GB in size, on top of the app itself. Each model’s download page shows its size; our “What AI Can My Computer Run?” tool lists the file size of each model along with whether it fits in your computer’s memory.
Where are downloaded models stored?
Ollama keeps models and its settings in a .ollama folder in your home folder (%HOMEPATH%\.ollama on Windows, ~/.ollama on a Mac). On Windows you can move them by setting the OLLAMA_MODELS environment variable. In LM Studio you can change the models folder from My Models.
Which model should I download first?
Ollama’s quickstart uses gemma4:e2b. Whether a model runs well depends on your computer’s memory: pick your graphics card or Mac in “What AI Can My Computer Run?” to see which models fit.
Where these steps come from
Every step and requirement was written from Ollama’s and LM Studio’s own documentation, linked beside each step. Every day we re-read those pages automatically and confirm the wording each step relies on is still there; they were last confirmed on Oct 3, 2026. If a page changes, the step is marked here until we’ve re-read it and updated the guide.
The pictures beside each step are simplified drawings, not screenshots, so the apps may look a little different. If a step doesn’t match what you see, the linked page is the up-to-date version; tell us and we’ll fix the guide.