This guide shows Windows users how to install Ollama and run qwen3:235b locally.

• OS: Windows 10 or later
• Disk: at least 220 GB free (model is about 142 GB, plus cache and headroom)
• RAM: 64 GB+ recommended (less can work, but it will be much slower)
• GPU: Optional, but a modern NVIDIA GPU helps a lot
• Stable internet for large model download

qwen3:235b is very large. If your PC is not high-end, it can still run but response speed may be slow.

Open PowerShell as Administrator and run:

irm https://ollama.com/install.ps1 | iex

1. Open: https://ollama.com/download/windows
2. Download OllamaSetup.exe
3. Run the installer

Open a new PowerShell window and run:

ollama --version

If you see a version number, install is done.

ollama pull qwen3:235b

This can take a long time because the model is large.

ollama run qwen3:235b

Type your prompt and press Enter.

Exit chat with:

/bye

ollama list
ollama ps
ollama show qwen3:235b
ollama stop qwen3:235b

Ollama runs a local API at http://localhost:11434.

$body = @{
  model = "qwen3:235b"
  messages = @(
    @{ role = "user"; content = "Give me a short summary of quantum computing." }
  )
} | ConvertTo-Json -Depth 5

Invoke-RestMethod -Uri "http://localhost:11434/api/chat" -Method Post -Body $body -ContentType "application/json"

• Expected on lower-end hardware for this model size
• Close heavy apps (browser tabs, games, IDEs)
• Keep enough free RAM and disk space

• Reboot Windows and retry
• Check model status:

ollama ps

• Re-pull model if download was interrupted:

ollama pull qwen3:235b

• Restart terminal
• Reinstall Ollama from the official Windows installer

Try a smaller Qwen3 model first:

ollama run qwen3:8b

Then move to qwen3:30b, and finally qwen3:235b when hardware allows.

• Ollama Windows download: https://ollama.com/download/windows
• Ollama Qwen3 model page: https://ollama.com/library/qwen3
• Qwen3 235B model card: https://huggingface.co/Qwen/Qwen3-235B-A22B-Instruct-2507