This is an array of Ollama Generate nodes using qwen3-vl:8b-thinking model to pull specific descriptive elements from an image and then combining them into a single output text which will then store into a directory with the source image. I've been pulling a LOT of my very first AI stuff out of mothballs and reinterpreting them through krea2. This is helpful for creating a descriptive framework to work from. Needs: https://ollama.com/download/

then run in cmd/terminal/powershell: "ollama pull qwen3-vl:8b-thinking" The nodes should then work in comfyui.