Members of the Artificial Intelligence Division (AI) of X, XAI, and the owner of the platform, Elon Musk, have released their new family of Grok 3 language models, which will rely on the EntEp Search search engine and promises to overcome A GPT-4O of OpenAI in the AIM and GPQA tests for the doctoral reasoning.
The presentation of the new X model of X has taken place this Tuesday at five in the morning (8:00 p.m. on Monday in the US), in a retransmission that has lasted about an hour and in which it has been explained To the desalle the one considered “the most intelligent on the planet”, in the words of the owner of the platform, Elon Musk.
This improved version of Grok 2 includes the characteristics of its predecessor, with improvements in sections such as chat or reasoning, as well as the generation of images; And adds others, such as the ability to reflect on the mistakes he makes, in order to achieve logical coherence.
The X artificial intelligence division has also indicated that it planned to launch Grok 3 in 2024. However, a few more months have been taken to profile it, try new capabilities and make it a reasoning model “much more capable than Grok 2”, according to Musk said during the presentation.
This coincides with what he advanced a few days ago, when he said in X that he had completed the preventive phase of the model “with 10 times more processing capacity than Grok 2”. However, in the meeting they have suggested that “perhaps” have a capacity 15 times greater.
“With the mission of understanding the universe”
First, the owner of X has indicated that Grok and, more specifically, the family of Grok 3 models, which will hide their reasoning processes, is composed of Grok-3 Reasoning and Grok-3 Mini Reasoning, which responds more quickly , although sacrifices the precision of their answers for it, as it has suggested.
He also pointed out that this family has been created “with the mission of understanding the universe” and that, therefore, it is still “in a kind of beta”, since some of its abilities are still “irregular”, such as the Voice mode “Literally, in 24 hours you will see improvements,” added other members of the XAI team during the retransmission.
«We believe that having the best training model is not enough. The best should think as a human being. You have to contemplate all possible answers, self -evaluate and verify errors, ”added those responsible for Grok 3, who have indicated that they can solve tasks related to STEM disciplines.
Superior to GPT-4O
In this sense, XAI has also affirmed that Grok 3 surpasses GPT-4O in reference tests such as Math Olympiad (Aime), which evaluates the performance of a model in a sample of mathematics questions, and GPQA, that is, that, Evaluate models using PhD Biology, Biology and Chemistry.
Thus, Grok 3 Reasoning and Grok 3 Mini Reasoning can carefully “think” of problems, similar to reasoning models such as OPENAI and DEPEEEK OPENAI and R1. Also, the first exceeds the best version of O3-mini-High, at several reference points, such as Aime 2025.
Also, the new XAI reasoning models support a new function in the iOS and Android application called Deepsearch. Like Google and Openai’s proposals, which have a similar name, this search tool collects information from the Internet and the X application to offer an exhaustive summary according to the consultation made.
For X subscribers
Grok 3 will first reach the subscribers of the premium+ level of X, although other functions will be restricted to a new plan that XAI has called Supergrok. This unlocks additional reasoning consultations, Deepsearch and offers unlimited generation of images.
Finally, those responsible for Grok have indicated that in a few weeks the Grok 3 models will be available through the XAI Application Programming Interface (API), together with Deepsearch, and that they also plan to release the Grok 2 source code in The coming months.
#Elon #Musk #presents #Grok #version #chatgpt #rival
