Grok 4.7

Grok 4.7 is SpaceXAI’s new flagship model for coding and knowledge work, marketed as twice as fast and half the price of comparable models while being served at the same price and speed as Grok 4.6. It uses a larger base model and a longer reinforcement-learning run focused on harder, multi-hour tasks, which improves its ability to manage long contexts, verify its own work, and handle conversational workflows via native understanding of the Grok Bot harness. Benchmarks show strong price-performance on extended coding tasks (CursorBench 4.0) and improved results on professional knowledge-work tests like GDPval and AA Briefcase, placing it competitively with other frontier models for document creation, presentations, and domain-specific workflows.

The release emphasizes safety and dual-use handling: an entirely new safeguard stack yields top results on refusals and jailbreak resistance, with LatchBio biosafety scoring 62.4% and HackerBench v0.3 allowing only 3.3% of risky cyber prompts through while still permitting legitimate security work. Select cybersecurity partners receive invite-only red-team access for defense research. Grok 4.7 is available now in Cursor and Grok Build, via the Grok API, third-party coding harnesses, model routers and cloud platforms; pricing starts at $2 per million input tokens and $6 per million output tokens, with a faster, double-priced variant for higher throughput.