Fusion Quill uses a GPU to accelerate AI performance. If you don’t have a GPU, it will work but will be slower.

If you have a Nvidia GPU, you can update your GPU driver and install CUDA for faster LLM and Image Generation performance.

If you have a AMD/Intel GPU, you can update your GPU driver for better Image Generation performance with DirectML.

No. Fusion Quill uses DirectML Onnx version of SD Models and Llama.cpp GGUF models for Local LLMs. So no Python installation is needed.

AI models range from 400MB to 5GB in size. Depending on your Internet speed, it will take some time. We do not include any models in the Fusion Quill Setup, since it would make the setup size much larger.

Make sure you PC does not go into Sleep mode when downloading.

If you use only Open AI APIs, you don’t need to download any models for Text use.