xAI's Grok 4.7 scores 46, trails Claude and GPT-6

Elon Musk's xAI has introduced Grok 4.7, describing it as its most capable model yet for coding and knowledge work. According to the company, it is built on a larger base model, trained with longer reinforcement learning, and designed to better verify its own output. Those are the design claims; the independent numbers tell a more mixed story.

On the Artificial Analysis Intelligence Index (v4.3.2), which combines ten benchmarks, Grok 4.7 scores 46 and lands mid-pack, while Claude Fable 5.1 and GPT-6 lead with 53 each. The gap widens in agentic coding: on Terminal-Bench 4.0, Grok 4.7 hits just 26 percent, versus 60 percent for GPT-6 Astra and 55 percent for Claude Fable 5.1, and even the cheaper DeepSeek V4.1 Flash edges past it at 27 percent. The counterweight is price: at $2 per million input tokens and $6 per million output tokens, its rates sit closer to Chinese models than Western frontier models, probably for good reason.

That combination is what the reader may want to weigh. The article does not say whether two of its highest reasoning levels performing about the same reflects a tuning choice or a ceiling. On the evidence given, the value proposition appears to hinge on whether buyers are willing to trade agentic coding performance for a much lower per-token price. The model is available through the Grok API, Cursor, and Grok Build.