Grok can now comprehend images, even explain jokes to you
What's the story
Elon Musk's artificial intelligence (AI) venture, xAI, has added image understanding capabilities to its Grok AI model.
The advanced feature lets paid users on X upload an image, and have a Q&A session with the AI chatbot about it.
The news was shared by both an xAI employee and the official Grok handle on X.
Development
New features to be refined over time
Musk revealed that the newly integrated image understanding feature also allows Grok to understand jokes.
However, he mentioned that this capability is still in its infancy and will be refined over time.
This comes after the release of Grok-2 in August, as a model and chatbot only for premium users on X.
Information
Grok-2 recently gained image generation capabilities
Along with its release, Grok-2 has also gained image generation capabilities. This was enabled by integrating the FLUX.1 model from Black Forest Labs. Earlier, xAI had announced plans to bring multimodal understanding as part of Grok's experience on X and within the developer API.
Document comprehension
It may soon understand PDFs
Musk also hinted at Grok's ability to understand documents in the near future.
Responding to a user's feedback that Grok can't process photos or PDFs, Musk said, "Not for long. We are getting done in months what took everyone else years."
This indicates that xAI is working at a breakneck speed to improve Grok's capabilities and provide more advanced features.