Vision LLM Structured Output with Sparrow

I show how Sparrow UI Shell works with both image and PDF docs to process and extract structured data with Vision LLM (Qwen2) in the MLX backend.