[xAI Releases Grok 4.7: Parameter Scale Increased to 2.1 Trillion, Still Trails Claude in Benchmarks] Beating AI News Flash: On Monday, xAI officially released Grok 4.7, with the company claiming it offers 'significant improvements' over Grok 4.6 at the same price and speed. Grok 4.7, which had been delayed at least five times, is now available on the Grok app, Cursor, Grok Build, and xAI API. The parameter scale of Grok 4.7 has reached 2.1 trillion, a 40% increase from Grok 4.6's 1.5 trillion. Input and output costs are set at $2 and $6 per million tokens, respectively. xAI has also incorporated supplemental training data from SpaceX, including Starlink satellite telemetry data, manufacturing records, and engineering failure logs, aiming to enhance the model's reasoning capabilities for hardware and physical systems. In benchmark testing, Grok 4.7 ranked second in both GDPval and AA-Briefcase, scoring 1695 and 1657, respectively, trailing Claude Fable 5.1's scores of 1735 and 1678. In the electrical engineering benchmark EEBench, it also ranked second, behind only GPT-6 Astra. According to CursorBench 4.0, Grok 4.7's task costs are higher than those of GPT-6 Astra and Claude Sonnet 5, while its overall performance still lags behind Fable 5.1. Elon Musk previously stated that Grok 4.7's performance is roughly equivalent to Claude Opus 5.0 but has not yet reached the level of the updated Opus 5.1. He also revealed that Grok 4.8, 4.9, and Grok 5 are currently in development.