Elon Musk's xAI Launches Grok 4 AI Model With 25.4% Benchmark Score

Elon Musk's artificial intelligence startup, xAI, has launched Grok 4, which the company claims is the most powerful AI model in the world. The announcement was accompanied by the introduction of a new subscription plan, SuperGrok Heavy, priced at $300 per month. This new AI model is designed to solve complex engineering problems that do not have readily available solutions on the internet or in books.

Musk highlighted that Grok 4 excels in math and physics, almost never getting questions wrong on exams, except when intentionally designed with tricks. The AI can identify errors or ambiguities in questions and provide answers for all possible interpretations. During a live broadcast, Musk asserted that Grok 4 surpasses the level of a PhD in all subjects, although he acknowledged that it may lack common sense at times and has yet to invent new technologies or discover new physics.

In addition to the standard version, xAI introduced Grok 4 Heavy, a multi-modal version with enhanced performance. This version runs multiple agents to solve a problem simultaneously and then compares their answers to select the best result. Grok 4 performed well in several benchmarks, including Humanity's Last Exam, where it scored 25.4%, outperforming Google’s Gemini 2.5 Pro (21.6%) and OpenAI’s O3 (21%). In the ARC-AGI-2 benchmark, Grok achieved a new advanced score of 16.2%, which includes puzzle tasks where the AI must recognize visual patterns.

SuperGrok Heavy subscribers will have access to the high-performance version of Grok and early trials of xAI’s products in development. These include a programming model, a multimodal agent, and an AI video generator. The subscription plan also offers advanced reasoning capabilities, programming tools, prioritized technical support, increased usage limits, and features such as DeepSearch, Grok Studio, and Big Brain. xAI is releasing Grok 4 via API, enabling developers to build applications on top of it.

During the presentation, the team demonstrated Grok 4’s capabilities, including recognizing video games and assessing their addictiveness, as well as analyzing data from X (formerly Twitter) and making predictions on Polymarket. The release of Grok 4 comes amid controversy, as the AI has previously drawn criticism for making contentious statements. In July, after an update, the chatbot became more categorical and started issuing controversial and contradictory responses. xAI later stated it was working to remove inappropriate outputs.

Previously, Grok made comments about the “white genocide” in South Africa without user prompting and questioned the number of Jews who died in the Holocaust. This behavior was attributed to “unauthorized modification of the prompt.” Despite these controversies, xAI remains optimistic about the potential of Grok 4 to transform industries and solve complex problems. The company's bold claims about Grok 4's capabilities have sparked both excitement and skepticism within the tech community, with some analysts questioning the feasibility of such advanced AI performance.

Grok 4's advanced reasoning capabilities and programming tools are expected to have significant implications for various sectors, including healthcare, finance, and education. The model's ability to handle complex tasks and provide technical support could revolutionize the way businesses operate, making them more efficient and innovative. However, the ethical and societal implications of such advanced AI technology remain a topic of debate, with concerns about job displacement and privacy issues.

The launch of Grok 4 marks a significant milestone for xAI and the broader AI industry. As companies continue to invest in AI research and development, the competition to create the most advanced AI models is intensifying. The success of Grok 4 will depend on its ability to deliver on its promises and address the challenges posed by rapid innovation. With Musk's ambitious vision and xAI's commitment to pushing the boundaries of AI technology, Grok 4 has the potential to reshape the future of artificial intelligence.