- Meta introduces Imagine, a new AI image generation capability.
- This feature uses Meta’s AI model Llama 3 to deliver sharper, higher quality images.
- This feature is part of a broader announcement Meta made last week.
Meta wants to let users unleash their imagination with the help of AI models.
Meta announced this week a new image generation feature called Imagine This allows users to create images in real time. The company is rolling out the feature in beta to users of WhatsApp and the Meta AI website.
To use it, users simply enter the “Imagine” prompt followed by a description of the image they are trying to generate. Mehta said the image changes “every time you type a few letters.”
This feature is powered by Llama 3, Meta’s latest open source large-scale language model. The company says it creates “sharper, higher-quality” images and “improves the ability to include text in images.”
Once users create an image, Mehta said, “they can animate it, repeat it in a new style, or even turn it into a GIF.”
Meta’s image generator is one of several updates to its AI assistant that the company announced last week. The assistant can respond to specific requests, such as restaurant recommendations with “sunset views” or “vegan options,” provide users with inspiration for new home decor, or help explain difficult concepts. Now you can support studying for big tests.
Meta CEO Mark Zuckerberg called the company’s “big AI news” “pretty wild,” according to Axios. But Meta is still catching up with competitors like OpenAI and Anthropic. The company is still training a large model for Llama 3, and when released it will likely be on par with OpenAI’s GPT-4 and Anthropic’s Claude 3 Opus, both released in March.
Check out the imagination in action below.
