Install
Download the installer from ollama.com. It runs as a background service and auto-detects NVIDIA GPUs. No CUDA toolkit install needed.
powershell
# Download from https://ollama.com/download/windows
# Or winget:
winget install Ollama.Ollama
# Verify service
ollama --version
ollama listWhen to use WSL instead
Stick with native Ollama for quick chat and Open WebUI. Switch to WSL2 if you need llama.cpp custom builds, ExLlamaV2, or fine-grained CUDA control.
text
Native Windows Ollama:
✓ One-click install
✓ NVIDIA GPU acceleration
✓ OpenAI-compatible API at :11434
Use WSL2 instead for:
→ ExLlamaV2 / EXL2 quants
→ Custom llama.cpp flags
→ vLLM / TabbyAPIRelated guides
IntermediateEdge / Local10 min read
WSL2 + Ollama GPU Passthrough on Windows
Run Ollama with NVIDIA GPU acceleration inside WSL2 — the most reliable Windows path for local LLMs.
BeginnerEdge / Local9 min read
llama.cpp on Windows with CUDA
Build llama.cpp with NVIDIA GPU support on Windows 11 — the path of least resistance for PC gamers.
BeginnerEdge / Local8 min read
8GB GPU Starter Guide: 3060 / 4060 / 3070
The most common local LLM hardware tier — which models, quants, and context lengths actually fit in 8GB VRAM.
Deployment guides are educational. Each model is subject to its own license — read the official Hugging Face model card before downloading or deploying.