Google continues to update its older language models and after making Gemini Live more natural in the face of users’ voice, last week expanded the experience of Gemini 2.0 Flash To the generation of images and discover that, apart from offering a sublime experience, users are beginning to use it in an unexpected way: to remove the water mark from the photos.
The photographs of the payment image banks are protected by water marks They prevent them from being used freely. A monthly subscription or the payment by an image allows them to be removed in order to use photographs that usually have some type of value, either by the subjects or environment of the scene or the quality of the same.
The expansion of Gemini 2.0 Flash experience has made it clear that it is facing a powerful image generator from Prompts as for Image editionsince it is even able to completely remove the water marks from those photographs mentioned to leave the original image.
A controversial use for a powerful generator of images that he is able to create profile images of an animated character just by giving him the front and thus making it clear that Google has been able to achieve something sublime, but he has had to limit, since he has left the door open to the free use of images with copyright with water marks that protect them.
Gemini 2.0 flash is available through Ai Studio Google for free and is marked as an experimental function and not to enter into production at the moment. The only thing to go through the tools for Google developers and enter an interface that is not aimed at the final consumer, but it does offer that unheard of generation of images.
Gemini 2.0 Flash, Available in Google’s ai Studio, Is Amazing at Editing images with simple text Prompts.
It Also Can Remave Watermarks from images (and puts ITS OWN Subtle watermark in instead 🤣) pic.twitter.com/znhtqjst1z
– Tanay Jaipuria (@tanayj) March 16, 2025
Unlike Claude 3.7 from Anthropic and GPT-4O, which does not allow the elimination of water marks, Gemini 2.0 Flash these days ago has allowed this type of modifications, so that a few hours ago it was already impossible to remove those water marks from the photographs.
Does not mean that it is a spectacular image generator which is giving free rein to certain interesting creations and adding people in imposite spaces with their generative filling capacity capable of imagining those spaces. A facet at all simple and in which in other AI models it is not as effective as what was achieved by Google.
At least, in the examples that have been seen, it did not work at all well by eliminating certain water marks, especially those that are semi -transparent and others that cover large portions of images according to Techcrunch. So at least on this side it does not fully fulfill that function that some users have found these days.
