OpenAI · LIVEBENCH 2026-06-25

GPT-6.1 Sol Max Effort

Published benchmark results for this specific model configuration.

gpt-6.1-sol-max

Overall score81.6 / 100

Cost / successful task$0.142USD · benchmark workload

Weight accessNot reported

Where this configuration performs

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

Reasoning92.6

Coding80.4

Agentic Coding54.5

Mathematics96.8

Data Analysis82.7

Language90.1

Instruction Following74.2

Task-level results

theory of mind86.5zebra puzzle100.0spatial98.0logic with navigation86.0code generation80.3code completion80.4javascript63.6typescript40.0python60.0AMPS Hard98.0integrals with game100.0math comp97.1olympiad92.3consecutive events90.6tablejoin57.4tablereformat100.0connections100.0plot unscrambling84.4typos86.0paraphrase72.6simplify72.0story generation70.8summarize81.3

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