Anthropic · LIVEBENCH 2026-06-25

Claude 4.7 Opus Thinking xHigh Effort

Published benchmark results for this specific model configuration.

claude-opus-4-7-xhigh-effort

Overall score76.5 / 100

Cost / successful task$0.528USD · benchmark workload

Weight accessNot reported

Where this configuration performs

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

Reasoning87.2

Coding82.1

Agentic Coding50.7

Mathematics92.9

Data Analysis78.3

Language77.9

Instruction Following66.7

Task-level results

theory of mind80.8zebra puzzle98.0spatial100.0logic with navigation70.0code generation85.9code completion78.3javascript63.6typescript43.3python45.0AMPS Hard98.0integrals with game84.0math comp98.0olympiad91.4consecutive events89.5tablejoin47.2tablereformat98.0connections92.3plot unscrambling61.4typos80.0paraphrase65.8simplify65.5story generation69.4summarize66.3

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