OpenAI · LIVEBENCH 2026-06-25

GPT-6 Luna Max Effort

Published benchmark results for this specific model configuration.

gpt-6-luna-max

Overall score72.0 / 100

Cost / successful task$0.026USD · benchmark workload

Weight accessNot reported

Where this configuration performs

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

Reasoning81.8

Coding79.0

Agentic Coding51.2

Mathematics89.1

Data Analysis73.4

Language73.8

Instruction Following55.9

Task-level results

theory of mind73.1zebra puzzle96.0spatial96.0logic with navigation62.0code generation77.5code completion80.4javascript63.6typescript40.0python50.0AMPS Hard98.0integrals with game74.0math comp94.1olympiad90.4consecutive events70.5tablejoin49.6tablereformat100.0connections100.0plot unscrambling57.5typos64.0paraphrase49.7simplify55.4story generation63.3summarize55.3

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