xAI · LIVEBENCH 2026-06-25

Grok 4.7 xHigh

Published benchmark results for this specific model configuration.

grok-4.7-xhigh

Overall score77.4 / 100

Cost / successful task$0.718USD · benchmark workload

Weight accessNot reported

Where this configuration performs

Category averages from the same release, on a 0–100 scale.

Reasoning82.7

Coding77.2

Agentic Coding54.0

Mathematics95.7

Data Analysis76.9

Language80.1

Instruction Following75.3

Task-level results

theory of mind84.6zebra puzzle86.0spatial94.0logic with navigation66.0code generation76.1code completion78.3javascript63.6typescript43.3python55.0AMPS Hard98.0integrals with game96.0math comp97.1olympiad91.7consecutive events85.7tablejoin44.9tablereformat100.0connections98.0plot unscrambling72.4typos70.0paraphrase78.0simplify70.0story generation74.9summarize78.2

Source files: Scores ↗ Categories ↗ Costs ↗ . Missing cost is unknown, never free. Compare configurations within the same release.