Qwen3.6-Max-Preview Benchmark Details

Qwen3.6-Max-Preview currently shows benchmark results led by MMLU Pro (5 / 126, score 88.50), LiveCodeBench (10 / 120, score 87.10), GPQA Diamond (16 / 177, score 90.40).

Benchmark Results

Qwen3.6-Max-Preview

Benchmark Results

综合评估

4 evaluations

Benchmark / mode

Score

Rank/total

GPQA Diamond

Thinking Level · Max

90.40

16 / 177

MMLU Pro

Thinking Level · Max

88.50

5 / 126

HLE

Thinking ModeTools

50.20

19 / 156

HLE

Thinking Level · Max

28.80

77 / 156

编程与软件工程

5 evaluations

Benchmark / mode

Score

Rank/total

LiveCodeBench

Thinking Level · Max

87.10

10 / 120

SWE-bench Verified

Thinking ModeTools

78.80

19 / 107

SWE-bench Multilingual

Thinking ModeTools

73.80

7 / 20

SWE-Bench Pro - Public

Thinking ModeTools

56.60

12 / 42

SWE-Bench Pro - Public

Deep Thinking ModeTools

57.30

10 / 42

指令跟随

1 evaluations

Benchmark / mode

Score

Rank/total

IF Bench

Thinking Level · Max

74.20

6 / 29

AI Agent - 工具使用

2 evaluations

Benchmark / mode

Score

Rank/total

Terminal Bench 2.0

Thinking ModeTools

61.60

16 / 46

Terminal Bench 2.0

Deep Thinking ModeTools

65.40

11 / 46

数学推理

1 evaluations

Benchmark / mode

Score

Rank/total

IMO-AnswerBench

Thinking Level · Max

83.80

10 / 19

生产力知识

1 evaluations

Benchmark / mode

Score

Rank/total

GDPval-AA

Deep Thinking Mode

12 / 21

Compare with other models