Developer Theo tests xAI's new Grok 4.6 model against Claude, GPT, and Kimi on coding, design, and long agent tasks, with mixed results.
Continue to AI University →