Meta’s Real-Time AI Image Generation: Revolutionizing WhatsApp

MetaversePlanet May 10, 2024

0 1 minute read

Meta is taking a bold step forward in artificial intelligence by introducing real-time AI image generation to WhatsApp. This feature is currently being rolled out as a beta for WhatsApp users in the US.

When users start typing a text-to-image prompt in a chat with Meta AI, they can see the image evolve in real-time as they add more details to their description of the desired creation.

Meta AI image feature

Meta has announced an exciting new feature for WhatsApp: real-time AI image generation. This feature, currently in beta for U.S. users, allows you to see an image evolve as you type in details. For instance, if you enter the prompt “Imagine a football match on Mars,” the image will dynamically update from a typical football player to a full football field on the Martian landscape. You can experience this by starting a chat with Meta AI on WhatsApp and using the prompt “Imagine.”

Moreover, Meta has enhanced its Meta Llama 3 model, which now produces sharper and higher-quality images and displays text more effectively. Additionally, you can ask Meta AI to animate the images you create, turning them into GIFs to share with friends.

This real-time image generation feature is also available on the web for U.S. users. It is part of a broader rollout of AI capabilities across all Meta platforms, including WhatsApp, Instagram, Facebook, and Messenger. The Meta AI assistant, introduced last September, is now integrated into the search boxes of Instagram, Facebook, WhatsApp, and Messenger and will soon appear directly in the main Facebook feed. It’s also accessible via an independent website, Meta.ai.

With the introduction of Llama 3, Meta aims to compete with existing AI models like ChatGPT. Llama 3, described as the next major version of Meta’s core open-source model, reportedly outperforms other models in its class on key benchmarks and excels in tasks such as coding.