Musk says new 2T param Grok model to be trained on SpaceX engineering corpus (twitter.com)

Elon Musk announced that SpaceX’s extensive engineering data corpus will soon enhance the training of the 2T param Grok model, excluding any ITAR-restricted material. This integration is expected to significantly bolster Grok’s engineering capabilities, making it a more powerful tool for complex tasks in the engineering domain. The supplemental training aims to fine-tune Grok’s performance and accuracy, particularly in comparison with existing models like Codex. This development is notable for the AI/ML community as it signifies a shift toward utilizing domain-specific data to train AI models, thereby enhancing their relevance and effectiveness in specialized fields. Rigorous side-by-side testing has already shown Grok Build’s superior accuracy with less dependence on user input, replacing Claude in tasks that require nuanced understanding. By leveraging real-world engineering challenges, Grok aims to establish itself as a trusted assistant in technical environments, showcasing the potential of tailored AI models in advancing industry-specific applications.