OpenAI is testing a “watermark” model for Image Generation, which is part of the ChatGPT model 4o.
This is a very interesting move. It’s probably because more users are creating Studio Ghibli artwork by using the ImageGen model.
ChatGPT has been in the news for a long time, largely because of the Image Generation model. This is the most advanced multimodel shipped to date.
Source: BleepingComputer
It can not only accurately generate images, but also realistic visuals. For example, you can create art like that produced by Studio Ghibli – a big name in Japanese studios.
ChatGPT’s ImageGen was previously only available to those who paid for the service (ChatGPT plus customers), but now it is available to all users, even those with a complimentary subscription.
As you can see, OpenAI appears to be working on”ImageGen” a new watermark for users who are not paying.
According to my sources, OpenAI has recently begun testing watermarks on images created using ChatGPT’s free account.
If you subscribe to ChatGPT plus, you will be able save images without watermark.
It’s not clear if OpenAI plans to watermark the images. OpenAI’s plans are always subject to changes.
ImageGen is an advanced model, API will be available soon
In a OpenAi’s Blog Post noted that the model was trained on a large number of images and text, which is why image generation is not just beautiful but also useful.
Open notes”We trained our models on the joint distribution of online images and text, learning not just how images relate to language, but how they relate to each other,” .
“Combined with aggressive post-training, the resulting model has surprising visual fluency, capable of generating images that are useful, consistent, and context-aware.”
Openaihas also confirmed that it is working on an ImageGen interface, so developers can create their own products.
