Elon Musk has confirmed that the Grok 5 AI model will be trained using a proprietary platform written in the C programming language by SpaceX. The primary goal of this system is to achieve the theoretical maximum performance of the hardware and increase model training speeds several times over. As reported by Ixbt.com reports .
Previously, there were reports that the SpaceX team was finalizing the first version of a training system written in C that interacts directly and as closely as possible with server hardware. This technology is a low-level solution that communicates directly with a cluster consisting of approximately 220,000 NVIDIA GB300 accelerators and 800G network interfaces.
As a reminder, in August last year, Elon Musk stated that xAI would launch the Grok 5 model by the end of the year and that it would be "amazingly excellent." The project timeline was later moved to the first quarter of 2026, though no specific dates have been announced yet.
According to developers, the Grok 5 model has the potential to reach the level of Artificial General Intelligence (AGI). This is expected to be a significant step toward AI matching human cognitive abilities.
Read “Zamin” on Telegram!
