Meta releases a web application that allows you to use the image generation AI 'Imagine' for free, allowing you to generate 4 images for each prompt using the Emu model

Meta has released a web application `` Imagine With Meta AI '' that allows you to use the image generation AI ``Imagine'' for free. Imagine With Meta AI is available for free and can generate four images just by entering a sentence (prompt). Overseas media have already published examples.

What's New Across Our AI Experiences | Meta

https://imagine.meta.com/

On the top page of Imagine With Meta AI, there is a button called ``Log in and generate'' that seems to be able to generate an image, but at the time of writing the article, even if you click the generate button, it says ``Unavailable in your area.'' was displayed, making it impossible to use in Japan.

Ars Technica, an overseas media, has published the image generation results with Imagine With Meta AI, so I picked up some of them. Below is an image generated with the prompt 'A cat in a car holding a can of beer'. We are able to generate high-quality images that follow the prompts. However, there are some slight anomalies, such as the unrealistic seat angle and the unrealistic angle of the right side window.

'A flaming cheeseburger' looks like this.

Below is an image generated with the prompt 'a photorealistic mickey mouse on the moon in a spacesuit'. Although it does not look like a live-action image, it produces a high-quality image of Mickey Mouse.