Elon Musk unveils Grok 4, following a controversy about antisemitism: Explore subscription costs and further details
Elon Musk's artificial intelligence (AI) company, xAI, has recently introduced Grok 4, a large language model that is being touted as one of the most advanced and capable AI systems currently available. Grok 4 boasts several advantages and differentiating features compared to OpenAI's ChatGPT and Google's Gemini models.
Grok 4's key capabilities include advanced multimodal and reasoning skills, enabling it to process both text and images, solve complex mathematics problems, generate images, and even predict future events, such as sporting outcomes. The AI model is claimed to surpass PhD-level expertise across all academic disciplines, achieving perfect SAT scores and being "smarter than almost all graduate students in all disciplines, simultaneously."
On key benchmarks, Grok 4 outperforms its competitors. For instance, on Humanity’s Last Exam (no tools), Grok 4 scored 25.4%, higher than Google’s Gemini 2.5 Pro (21.6%) and OpenAI’s o3 (21%). In visual reasoning tasks, such as ARC-AGI-2, Grok 4 scored 16.2%, nearly doubling the performance of Claude Opus 4 and significantly outpacing Google’s Gemini 2.5 Pro and OpenAI’s models (~8% and ~6% respectively).
xAI also offers Grok 4 Heavy, a multi-agent version that enables parallel processing and even higher performance on complex tasks. With tools, Grok 4 Heavy scored 44.4% on the Humanity’s Last Exam—substantially ahead of GPT-4 with tools (26.9%) and Gemini 2.5 Pro (21% with tools).
Grok 4 also features enhanced voice capabilities and is being integrated into Tesla vehicles, opening new pathways for conversational AI in automotive experiences. The model is expected to support images, videos, and text like OpenAI's GPT-5o and Google's Gemini 2.5 Pro.
In comparison to ChatGPT and Gemini, Grok 4 leads on both general intelligence and visual reasoning, as demonstrated by its outstanding scores on Humanity’s Last Exam and ARC-AGI-2. Grok 4’s multimodal functionality, multi-agent processing (via Grok 4 Heavy), and integration with Tesla vehicles give it an edge over competitors, who are mainly focused on cloud and API-based services.
The standard subscription plan for Grok 4 starts at $30 per month. Users can also opt for the SuperGrok Heavy plan at $300 per month.
In summary, Grok 4 is marketed—and benchmarked—as a leading-edge AI model with superior academic reasoning, multimodal skills, and unique deployment advantages, particularly in automotive AI, compared to ChatGPT and Gemini.
The advanced multimodal AI system, Grok 4, developed by Elon Musk's xAI, leverages technology to process both text and images, outperforming competitors like ChatGPT and Google's Gemini in areas such as general intelligence and visual reasoning. Moreover, Grok 4's integration with Tesla vehicles and availability in subscription plans starting at $30 per month make it a significant contender in the realm of artificial intelligence, particularly in the automotive sector.
