It should be possible get a description of an image using AI. The image should be sent to a LLM in Ollama and a description of the image should be provided.

This feature would help blind people to interact with the text.

UX

• Add a button to top right corner of the image.
• If the user clicks, a description of the image is streamed bellow the image.

Implementation

Take the image from the asset folder and send to Ollama. Stream the answer from Ollama to the text editor. Example: https://ollama.com/blog/vision-models.

This feature won't be much different of other AI features of this project. Check the interactions with OllamaApi class.