Vision-Language SFT This skill assumes already routed here: the data shape is image+text demonstrations, not preference pairs or a verifiable reward signal, and the base is a vision-language model rather than a text-only one. covers the text-only LoRA/QLoRA recipe this skill specializes for the vision tower and projector; read that skill first if the LoRA fundamentals (rank, alpha, target modules) aren't already familiar. Input: an image+text dataset and a VLM base model already picked from the model catalog. Output format: a validated adapter config — which components are frozen, LoRA target…