xAI adds image understanding capabilities to Grok

Elon Musk-owned xAI has added image-understanding capabilities to its Grok AI model. With that, paid users on X social platform can upload an image and ask the AI chatbot questions about it.

One of the xAI employees and the official Grok handle posted about this update on X.

In a separate post, Musk said that Grok can even explain the meaning of a joke through the new image understanding feature. He added that the functionality is in the early stages and will improve along the way.

In August, Musk’s AI company released Grok-2 as a model and in the form of a chatbot for premium users on X. The chatbot on the social network also gained image-generation capabilities using FLUX.1 model by Black Forest Labs.

At that time, xAI said that it would release multimodal understanding as part of Grok’s experience on X and the developer API.

Grok might also understand the documents soon. In a reply to a user’s feedback about Grok not being able to handle photos or PDFs, Musk said, “Not for long. We are getting done in months what took everyone else years.”

The social network has been trying to add more features to the AI chatbot and the paid user tiers to make the offering more attractive. Earlier this month, X rolled out a new tool called Radar for Premium+ subscribers to observe real-time trends and provide insights into conversations.