It appears that floodgates have opened in the final month of 2023, as tech firms are launching new AI models and tools virtually left, right, and center. Yesterday, Google unveiled its largest AI model called Gemini, and shortly before, Apple also released its machine learning framework MLX. Now, Meta has also launched its AI image generator tool called Imagine with Meta AI, which was so far only available in chats as a standalone platform, allowing anyone with a Meta account to generate stunning AI art. For now, the service is only available in the US, but it is expected to be expanded to other regions soon.
Meta announced its launch in a blog post, where it said, “Today, we’re expanding access to imagine outside of chats, making it available in the US to start at imagine.meta.com. This standalone experience for creative hobbyists lets you create images with technology from Emu, our image foundation model”. Even after its standalone launch, the feature is still available in Meta messaging platforms too.
Imagine with Meta AI is a text-to-image AI model that can generate images similar to DALL-E, Midjourney, and Stable Diffusion. It is powered by Meta’s existing technology called the Emu image generation model and is capable of creating high-resolution images from text prompts. The platform is free-to-use, unlike the others mentioned here which makes it a unique proposition for users.
A thing to note here is that Meta has said it will be adding watermarks in weeks to come. The statement said, “In the coming weeks, we’ll add invisible watermarking to the Imagine with Meta AI experience for increased transparency and traceability”. Unlike normal watermarks, it will be invisible to the eye but will show up in detection tools. This special watermarking will be used because it is resilient to common image manipulations like cropping, color change (brightness, contrast, etc.), screenshots, and more.
If you have used this feature in Meta chats, then you already know the basics. In case, you have not used it before, this is how it will work. You first need to go to the official website, which is mentioned above. Then, you will have to log into Meta’s account to be able to generate images. Once done, you will see a text field where you can enter your prompt in natural language, which will then be processed by the AI platform using Natural Language Processing (NLP). Each prompt will generate four different images per iteration.
Imagine with Meta is not the only tool announced by the company in its long announcement post. Another interesting feature, which will be available only in the Messenger app, is Riff on Images. This feature allows your friends in a group chat to add more elements to an initial AI-generated image to pass it back and forth for entertainment and hilarity.
Reels in Meta AI chats is another feature coming to Messenger. It is an AI-powered recommendation tool that is being embedded within Meta AI. It can now share Reel recommendations on a particular topic when asked. These features have already started being rolled out and should reach to you soon, as long as you live in the US.
Copyright © HT Media Limited
All rights reserved.