Scoop: SpaceXAI launches new model, Grok 4.5
TL;DR
SpaceXAI is launching Grok 4.5 as its first major model since going public and acquiring Cursor. The pitch is coding, agentic workflows and knowledge work, not a broad consumer chatbot. The company says Grok 4.5 was trained closely with Cursor and beats comparable models on engineering and knowledge-work benchmarks. Those benchmark charts come from SpaceXAI’s own announcement.
Nauti's Take
This is a developer launch, not a consumer moment. With Cursor, SpaceXAI has bought its way into one of the most valuable surfaces in AI: the place where code actually gets written.
The performance claims are intentionally aggressive and still PR-heavy until independent tests land. Grok 4.5 becomes interesting if it can plan, patch and test reliably in real repositories while staying cheaper than Claude or OpenAI.
EU users are watching from outside at launch.
Briefingshow
Grok 4.5 shows where the model market is moving: away from standalone chatbots and toward tools embedded in coding environments and agent workflows. The important part is not just the benchmark story, but distribution through Cursor. If developers can try the model inside their editor without extra setup, product proximity can matter nearly as much as raw model quality.
Sources
