JHubi1
Does setting the tts service to Google's solve the issue?
Images are supposed with multimodal models. Take a look into `llava`
Did you follow the steps described in [the wiki](https://github.com/JHubi1/ollama-app/wiki/Components#multimodal-input)?
Please open a new issue, this isn't related
> gemma3:12b is multimodal but app does not allow to upload images or files Please see #83
Elaborate please
Could you go into the global Android settings and enable all permissions (two to be precise), then try again?
Is this still something that's being worked on?
Maybe you mean preloading? The app sends an empty request to the backend to load the model in memory
This is simply not true. Ollama App is still in active development. Developing a good app is simply not an easy task. And, if you didn't know already, most of...