Grok’s AI Supremacy: Mensa Scores Prove Its Excellence Over GPT and Gemini
In the ever-evolving landscape of artificial intelligence, Grok-4 Expert Mode has emerged as a formidable contender, securing top positions in the latest IQ test rankings by AItracking.org. With an impressive 126 score offline and a remarkable 136 on Mensa Norway, Grok-4 Expert Mode has firmly established its dominance in AI intelligence leaderboards.
The Vision variant of Grok also showcases its versatility, achieving a score of 74 offline and 96 on Mensa Norway. This highlights the steady progress in multimodal capabilities, a challenging yet crucial aspect of AI development. The comparison chart, with its black bar for Mensa Norway and orange for offline IQ-style evaluation, provides a clear visual representation of Grok’s prowess.
Google’s Gemini 3 Pro Preview, a close contender, follows just below Grok, with scores of 123 offline and 142 Mensa. OpenAI’s GPT-5.1 family is not far behind, with the Pro version crossing 118 offline and 144 Mensa. The Thinking and Vision models within the GPT-5.1 family demonstrate impressive consistency, falling within the 95-120 range.
Claude, renowned for its reasoning abilities, maintains a strong position with its 4.5 Opus and Sonnet models, scoring between 114-124 offline and 121-124 Mensa. However, it is Grok that takes the spotlight, showcasing its exceptional capabilities.
As the rankings descend, models like Perplexity, DeepSeek R1, Mistral Medium 3.1, and Llama 4 Maverick cluster around the 88-99 offline and 95-107 Mensa range, solid performers but a tier below the top contenders. Earlier-generation models, such as GPT-4o, Grok-4.1 Beta, and Llama 4 Vision, occupy the lower ranks, with scores in the 60-70 offline and 67-96 Mensa range, underscoring the rapid pace of AI advancement.
The article concludes by highlighting a potential shift for Microsoft’s AI chatbot Copilot, which is set to exit WhatsApp in January 2026. This development prompts users to consider the changes they might witness in their interactions with AI chatbots.