Anthropic · LIVEBENCH 2026-06-25

Claude 5.5 Opus Thinking xHigh Effort

Published benchmark results for this specific model configuration.

claude-opus-5-5-xhigh-effort

Overall score82.1 / 100

Cost / successful task$0.243USD · benchmark workload

Weight accessNot reported

Where this configuration performs

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

Reasoning90.7

Coding89.3

Agentic Coding65.4

Mathematics96.8

Data Analysis79.8

Language85.5

Instruction Following67.0

Task-level results

theory of mind84.6zebra puzzle100.0spatial98.0logic with navigation80.0code generation91.5code completion87.0javascript72.7typescript53.3python70.0AMPS Hard99.0integrals with game99.0math comp97.1olympiad92.2consecutive events90.4tablejoin51.0tablereformat98.0connections99.3plot unscrambling77.2typos80.0paraphrase64.7simplify63.0story generation71.2summarize69.3

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