Parsa

@autmoate @geoffroy-noel-ddh Sorry to hear that you're running into issues. Windows requires some additional steps w/ the latest revision - do you have FFMPEG and Pyvips installed to your machine?...

> How to run in Android device ? with this VLM can i create standalone app without any server dependency ? Soon you will be able to run this locally...

> I saw in the video demo that the file coordinates were multiplied by the image size, which seems to work. Correct - you multiply file coordinates by image size...

> I also noticed that the results were different, which is why I tried to set up moondream2 with transformers, because the model here: [huggingface moondream2](https://huggingface.co/vikhyatk/moondream2) is larger and I...

> Many thanks for the reply! [@parsakhaz](https://github.com/parsakhaz) Good to know and sure, longer time for inference on CPU is normal. > > Nevertheless, the question remains why the results differ,...