The Case for Cursor’s Acquisition

On Friday, SpaceX completed its $60 billion acquisition of Cursor two months after announcing the deal. In a blog post announcing the acquisition, Cursor said the rationale for the merger was to give it access to SpaceX’s compute, letting it build “stronger models” that are also more cost-effective than it could on its own. SpaceX released Grok 4.6 on August 12, its second joint model with Cursor following Grok 4.5 in July, and independent benchmarker Artificial Analysis scored it 61 on its Intelligence Index. In other words, with SpaceX’s acquisition of Cursor, a company that could not compete on coding a year ago is now at the frontier.

For SpaceX, the deal lets it take further advantage of its unique edge among the various players in the AI race, which is the ability to build compute quickly at scale. xAI built Colossus, the largest single-site-scale compute cluster, in a matter of weeks. However, the company has struggled to match other frontier labs in performance and adoption, with Elon saying in March the company was not "built right the first time around” and had to be rebuilt from the ground up. Cursor had the inverse problem. It described itself as compute-starved while building competitive coding models on open-weight bases through heavy post-training. Grok 4.6's gains came almost entirely from post-training rather than a larger base model, an early indication of how the acquisition may benefit both parties.