"Despite the apparent simplicity of the challenge, Gemini was the only LLM that was able to create a web site that rendered well on mobile. Grok came in second with a mobile rendering that featured odd spacing and horizontal lines placed seemingly randomly but at least kept the text legible. Both ChatGPT and Claude’s styling resulted in mobile rendering that can only be described as broken with multiple instances of overlapping text."
How is the AGI doing Jensen Huang?