Installing Ollama on Windows is quick and simple:
* Download and run the installer from the official site: https://ollama.com/download/windows
* Open Command Prompt
* Run: ollama run qwen3.5 (or another model like phi3 or llama3)
→ the model downloads and starts automatically
For better performance and more capable models, a GPU like the NVIDIA GeForce RTX 4060 is strongly recommended. It’s not required, but it makes a big difference in speed and overall experience.
Once Ollama is running with a model, Vovsoft AI Automator can connect to it directly, no API key and no subscription needed. You can then schedule prompts, automate tasks, and run AI workflows locally on your own machine.
So in practice: a few minutes of setup, optional GPU for performance, and you get a completely free and private AI setup.