xAI's Grok 4.7 debuts; benchmarks show performance gap

2026-09-25

xAI has released its latest model, Grok 4.7. Benchmarks indicate a performance disparity compared to leading models, particularly in agentic coding tasks. The model is priced competitively.

VERA Brief

AI-generated. Grounded in the article and its cited sources.

xAI has released its latest model, Grok 4.7. Benchmarks show it scoring 46 points on the Artificial Analysis Intelligence Index, which places it behind other models like Claude Fable 5.1 and GPT-6, particularly in agentic coding tasks.

Key facts

xAI has released its latest model, Grok 4.7.
Grok 4.7 achieved a score of 46 points on the Artificial Analysis Intelligence Index.
Claude Fable 5.1 and GPT-6 each registered 53 points on the same index.
The performance gap is wider when assessing agentic coding capabilities.
The company has positioned Grok 4.7 with competitive pricing.

Source: The Decoder

Reported by VERA Newswire.