OpenAI · LIVEBENCH 2026-06-25

GPT-5.6 Terra Max Effort

Published benchmark results for this specific model configuration.

gpt-5.6-terra-max

Overall score77.9 / 100

Cost / successful task$0.352USD · benchmark workload

Weight accessNot reported

Where this configuration performs

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

Reasoning90.6

Coding78.2

Agentic Coding54.9

Mathematics94.9

Data Analysis79.3

Language82.9

Instruction Following64.6

Task-level results

theory of mind86.5zebra puzzle100.0spatial100.0logic with navigation76.0code generation76.1code completion80.4javascript68.2typescript46.7python50.0AMPS Hard98.0integrals with game95.0math comp95.1olympiad91.5consecutive events87.7tablejoin50.3tablereformat100.0connections93.0plot unscrambling69.7typos86.0paraphrase57.0simplify61.4story generation66.8summarize73.2

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