Development Windows

Ollama for Windows

Run open AI models on your Windows PC, chat from the terminal and connect applications through a local API with Ollama.

WindowsFree and open-source software (MIT); cloud usage has separate pricing.Version 0.40.12
Operating system
Windows 10 22H2 or newer, Home or Pro
Architecture
Not specified for the EXE on the publisher download page
Download package
EXE installer (OllamaSetup.exe)
Download Ollama for Windows

GPU acceleration requires compatible hardware and drivers. At least 4 GB for the application; models need additional storage. The Windows guide separately lists an amd64 standalone CLI archive.

Link checked · Download guidance

What is Ollama?

Ollama brings open AI models to a Windows PC for local chat and application development. Use the ollama command in Command Prompt or PowerShell to download a model and start a conversation, or connect your own application to the local API. Local requests do not require an API key. Cloud models are also available, but they run on Ollama’s servers rather than your computer.

The native Windows application runs in the background and supports compatible NVIDIA and AMD Radeon GPUs. The publisher lists Windows 10 22H2 or newer, Home or Pro, and at least 4 GB of space for the application. Model files need additional storage and can occupy tens to hundreds of gigabytes. Response speed depends on the model and your hardware; a large model can be slow without a strong GPU.

The linked EXE is the installer offered by the publisher’s Windows download page. At verification, it resolved to the v0.40.12 release. The screenshot shows Ollama’s Windows command-line interface, not a demonstration of the current desktop interface.

What it offers

  • Download an open model and start a local terminal chat with ollama run.
  • Connect applications to the local API at localhost:11434 without a cloud API key.
  • Run natively on Windows with support for compatible NVIDIA and AMD Radeon GPUs.
  • Choose local models or sign in to use cloud models.

Good to know

  • Local model speed depends on hardware; large models can be slow without a strong GPU.
  • The application needs at least 4 GB of disk space, plus separate model storage that can reach tens to hundreds of gigabytes.
  • Cloud models run remotely and require usage credits or a paid plan; they are not local execution.

Product images

Sources

References used for this listing. Retrieval dates do not indicate hands-on testing.

  1. Ollama - Download ↗Retrieved
  2. Download Ollama on Windows ↗Retrieved
  3. Ollama documentation — Windows ↗Retrieved
  4. Ollama documentation — Quickstart ↗Retrieved
  5. Ollama source license ↗Retrieved
  6. Ollama v0.40.12 — publisher release metadata ↗Retrieved

Browser storage

No analytics or advertising cookies. If you dismiss this notice, we remember its acknowledgement in your browser for 180 days. How storage works