After the emergence of GPT-5, DeepSeek has once again pressured the delayed Grok 3
Elon Musk stated that Grok 3 is "scarily smart," and its performance in testing surpasses any released product we know of. We have entered the final stage, and Grok 3 will be released in about one to two weeks
The battle of AI large models continues. Following DeepSeek's shock in the AI circle, OpenAI has "hinted" at the launch of GPT-5, and Musk has also announced that Grok 3 may be released in the next week or two.
On Thursday, at the 12th World Government Summit (WGS 2025) held in Dubai, Musk stated that Grok 3 is "scarily smart":
In testing, Grok 3's performance surpasses any released product we know of. We are in the final stages, and Grok 3 will be released in about one to two weeks.
It is worth mentioning that the Grok 3 model will be launched by Musk's artificial intelligence company xAI, which is expected to introduce "chain of thought" reasoning capabilities, allowing it to process complex tasks step by step like human cognitive processes.
Last summer, Musk posted on social platform X that xAI's next-generation AI model Grok 3 would be trained on 100,000 H100 GPUs and would be launched by the end of 2024, stating it "will be very special." However, it is now 2025, and Grok 3 has yet to be released.
Analysts suggest that the small size of the xAI team may also be one of the reasons for the delay in Grok 3's release. Musk stated in an interview that when asked whether Grok 3 would reach industry-leading levels, he replied, "Hopefully, that's our goal."