OpenAI · LIVEBENCH 2026-06-25

GPT-5.5 Thinking xHigh Effort

Published benchmark results for this specific model configuration.

gpt-5.5-xhigh

Overall score80.2 / 100

Cost / successful task$0.435USD · benchmark workload

Weight accessNot reported

Where this configuration performs

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

Reasoning89.7

Coding82.1

Agentic Coding54.0

Mathematics95.9

Data Analysis81.6

Language87.4

Instruction Following70.7

Task-level results

theory of mind84.6zebra puzzle100.0spatial98.0logic with navigation76.0code generation81.7code completion82.6javascript63.6typescript43.3python55.0AMPS Hard97.0integrals with game99.0math comp96.1olympiad91.4consecutive events88.8tablejoin55.9tablereformat100.0connections100.0plot unscrambling74.1typos88.0paraphrase71.7simplify66.3story generation72.5summarize72.3

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