OpenAI · LIVEBENCH 2026-06-25

GPT-6 Sol Max Effort

Published benchmark results for this specific model configuration.

gpt-6-sol-max

Overall score79.2 / 100

Cost / successful task$0.268USD · benchmark workload

Weight accessNot reported

Where this configuration performs

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

Reasoning88.7

Coding81.8

Agentic Coding52.9

Mathematics96.4

Data Analysis81.2

Language85.3

Instruction Following68.6

Task-level results

theory of mind84.6zebra puzzle100.0spatial96.0logic with navigation74.0code generation83.1code completion80.4javascript63.6typescript40.0python55.0AMPS Hard98.0integrals with game99.0math comp97.1olympiad91.4consecutive events89.5tablejoin54.0tablereformat100.0connections100.0plot unscrambling77.9typos78.0paraphrase70.1simplify69.3story generation66.9summarize68.0

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