Iris should be able to process things beyond text files, depending on the supporting model.
- png, jpeg, etc
- pdf
- docx
- etc
- Add an 'attachment' 📎 button somewhere near the input box.
- Allow for drag & drop of files into the chat window
- If possible, filter file types by model capability, and surface incompatibilities with a reasonable error message, eg ("this model doesn't support vision")
Iris should be able to process things beyond text files, depending on the supporting model.