SpaceXAI launched Grok 4.7 on Monday, September 21, calling it the company’s most powerful model yet for coding and knowledge work, available now in Cursor, Grok Build, and the Grok API.
The new model uses a larger base model than Grok 4.6 and was trained with a longer reinforcement learning run on a harder mix of tasks, weighted toward the Grok Bot harness. SpaceXAI says pricing starts at $2 per million input tokens and $6 per million output tokens, and the model scores 46.3 percent on CursorBench 4.0 for software engineering, ahead of Grok 4.6 at 40.4 percent and GPT-5.6 Sol Max at 41.7 percent.
The launch comes ten days after Musk said Grok 4.7 needed a few more days to cook, following his roadmap that runs through Grok 4.8 training and into Grok 5. SpaceXAI has been shipping Grok updates at a rapid clip, including Grok Voice Transcribe 2.0 last week.
SpaceXAI said Monday that Grok 4.7 works longer on difficult tasks, checks its work more carefully, and ships with the company’s strongest safeguards to date. The model also ranks third for agentic coding on the Artificial Analysis Intelligence Index with a score of 46, and Musk called it “a strong combination of intelligence, speed & low cost.”
SpaceXAI has not said when Grok 4.7 will reach the consumer Grok app or the free tier, and it is unclear whether the $2 and $6 per million token pricing will hold as capacity scales. For now, the open question is how quickly Grok 4.7’s coding gains show up for everyday Grok users outside the API and developer tools.
Want to see more of our stories on Google?
P.S. — Buying a new Tesla? Click here to save $1,000 USD, while supporting independent news.
Help support us by shopping on Amazon here.
Links in this post are affiliate links, so we earn a tiny commission at no charge to you. Thanks for supporting independent media!
