Alibaba · LIVEBENCH 2026-06-25

Qwen 3.6 Plus

Published benchmark results for this specific model configuration.

qwen3.6-plus

Overall score68.9 / 100

Cost / successful task$0.227USD · benchmark workload

Weight accessNot reported

Where this configuration performs

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

Reasoning75.8

Coding78.2

Agentic Coding41.4

Mathematics83.7

Data Analysis69.9

Language75.0

Instruction Following58.3

Task-level results

theory of mind67.3zebra puzzle68.0spatial98.0logic with navigation70.0code generation80.3code completion76.1javascript59.1typescript20.0python45.0AMPS Hard97.0integrals with game62.0math comp93.1olympiad82.8consecutive events67.0tablejoin44.7tablereformat98.0connections90.5plot unscrambling52.5typos82.0paraphrase60.6simplify51.6story generation56.9summarize64.3

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