v1.1.0

Features

• Ollama describer will pull model from Ollama libary if absent. Also better messages for most common errors.
• the path of the image, relative to the given root is now saved in the answer files under meta.image.path gh-30
• improved detection of json output in reports
• json answers are now formatted to html in the report

v1.0.1

Tested and documented support for Ollama, ovis, qwen-vl, smol and moondream models.