Meta has launched a new tool that allows users to create images using artificial intelligence.
The tool, which the company calls “Visualize,” is based on its own image model (Emu).
The tool was previously available to users of the Facebook Messenger application, but Meta decided to launch it as an independent tool that can be accessed and used on the web via this link (imagine.meta.com).
Meta says it will add invisible watermarks to images created with the Imagine tool in the coming weeks and will expand it to other image generation products.
Watermarks are used to distinguish real or human-generated images from images created using artificial intelligence tools.
Meta has made some improvements to the Imagine tool in the Messenger app, allowing users to use additional commands to modify or "reimagine" images in response to images created with them.
The new profiler is similar to other similar tools such as DALL-E, Midjourney, and Stable Diffusion in that it creates high-resolution images based on user-entered text commands.
The company currently offers the tool for free and notes that it generates four different images for each text command the user enters.
The company also announced some improvements to its Meta AI assistant, which will provide detailed answers and accurate summaries, and short Reels videos will appear in Meta AI conversations.