Qwen3.7 Max Benchmark Details

Qwen3.7 Max currently shows benchmark results led by MMLU Pro (4 / 134, score 89.60), LiveCodeBench (5 / 128, score 91.60), GPQA Diamond (27 / 274, score 92.40). This page also compares it with 3 competitor models and 3 predecessor or same-series models, including performance and pricing views when available.

Benchmark Results

Qwen3.7 Max

Benchmark Results

General Knowledge

4 evaluations

Benchmark / mode

Score

Rank/total

MMLU Pro

Thinking Level · Max

89.60

4 / 134

LiveBench

Deep Thinking Mode

74.29

21 / 115

HLE

Thinking ModeTools

53.50

24 / 197

HLE

Thinking Level · Max

41.40

71 / 197

Other

1 evaluations

Benchmark / mode

Score

Rank/total

GPQA Diamond

Thinking Level · Max

92.40

27 / 274

Coding and Software Engineer

6 evaluations

Benchmark / mode

Score

Rank/total

Text Arena (Coding)

Standard Mode

1540.77

11 / 35

LiveCodeBench

Thinking Level · Max

91.60

5 / 128

SWE-bench Verified

Thinking ModeTools

80.40

13 / 116

SWE-bench Multilingual

Thinking ModeTools

78.30

7 / 29

SWE-Bench Pro - Public

Thinking ModeTools

60.60

15 / 62

SciCode

Thinking Level · Max

53.50

8 / 16

Common Sense Reasoning

1 evaluations

Benchmark / mode

Score

Rank/total

SimpleBench

Standard Mode

70.40

14 / 92

Instruction Following

1 evaluations

Benchmark / mode

Score

Rank/total

IF Bench

Thinking Level · Max

79.10

5 / 36

AI Agent - Tool Usage

2 evaluations

Benchmark / mode

Score

Rank/total

MCP-Atlas

Thinking ModeTools

76.40

18 / 41

Terminal Bench 2.0

Thinking ModeTools

69.70

5 / 48

Text Embedding

1 evaluations

Benchmark / mode

Score

Rank/total

Context Arena

Standard Mode

79 / 126

Math and Reasoning

1 evaluations

Benchmark / mode

Score

Rank/total

IMO-AnswerBench

Thinking Level · Max

3 / 24

Compare with other models

Competitor Comparison

Benchmark scores for Qwen3.7 Max compared against top models in its class

Qwen3.7 MaxClaude Opus 4.6 Kimi K2.6 GLM 5.1

Benchmark categories:

The chart shows each model’s highest score per benchmark within the current filter. Out-of-100 benchmarks use raw heights; out-of-range benchmarks are scaled within that benchmark while labels keep the original scores.

12 benchmarks with comparable scores. Each model shows its best score; mode label is displayed below.

Benchmark	Qwen3.7 MaxCurrent	Claude Opus 4.6	Kimi K2.6	GLM 5.1
HLE Accuracy 综合评估	53.50Thinking Enabled ｜ Tools	53.00Extended Thinking ｜ Tools	54.00Thinking Enabled ｜ Tools	52.30Thinking Enabled ｜ Tools
LiveBench Accuracy 综合评估	74.29Deep Thinking Mode	76.33Thinking Level · High	72.17Thinking Enabled	70.18Standard Mode
GPQA Diamond Accuracy 科学与综合推理	92.40Thinking Level · High	91.31Extended Thinking	90.50Thinking Enabled	86.20Thinking Enabled
LiveCodeBench Pass @K 编程与软件工程	91.60Thinking Level · High	76.00Extended Thinking	89.60Thinking Enabled	--
SWE-bench Multilingual Accuracy 编程与软件工程	78.30Thinking Enabled ｜ Tools	72.00Extended Thinking ｜ Tools	76.70Thinking Enabled ｜ Tools	--
SWE-Bench Pro - Public Accuracy 编程与软件工程	60.60Thinking Enabled ｜ Tools	--	58.60Thinking Enabled ｜ Tools	58.40Thinking Enabled ｜ Tools
SWE-bench Verified Accuracy 编程与软件工程	80.40Thinking Enabled ｜ Tools	80.84Extended Thinking ｜ Tools	80.20Thinking Enabled ｜ Tools	--
Text Arena (Coding) Arena Score 编程与软件工程	1540.77Standard Mode	1555.35Standard Mode	--	1534.00Standard Mode
SimpleBench Score (AVG@5) 常识推理	70.40Standard Mode	67.60Standard Mode	--	55.10Standard Mode
IF Bench Accuracy 指令跟随	79.10Thinking Level · High	94.00Extended Thinking	--	--
MCP-Atlas Pass rate / claim coverage AI Agent - 工具使用	76.40Thinking Enabled ｜ Tools	76.80Thinking Level · High ｜ Tools	69.40Thinking Enabled ｜ Tools	75.60Standard Mode ｜ Tools
Terminal Bench 2.0 Accuracy AI Agent - 工具使用	69.70Thinking Enabled ｜ Tools	65.40Extended Thinking ｜ Tools	66.70Thinking Enabled ｜ Tools	63.50Thinking Enabled ｜ Tools

2 additional benchmarks remain in the chart above.

Standard API Pricing: Qwen3.7 Max vs. Peer Models

Shows standard text input and output pricing side by side for each model. If extended-context pricing exists, the chart keeps the base rate and explains the threshold below.

Source: DataLearnerAI. Standard text prices shown here use the default supplier.

These models use different currencies or billing units, so the page falls back to raw price values instead of a shared bar chart.

Qwen3.7 Max

Supplier: 阿里巴巴

Standard input: ¥12 / 1M tokens

Standard output: ¥36 / 1M tokens

Claude Opus 4.6

Supplier: Anthropic

Standard input: $5 / 1M tokens

Standard output: $25 / 1M tokens

Kimi K2.6

Supplier: Facebook AI研究实验室

Standard input: $0.95 / 1M tokens

Standard output: $4 / 1M tokens

GLM 5.1

Supplier: 智谱AI

Standard input: $1.4 / 1M tokens

Standard output: $4.4 / 1M tokens

Model	Supplier	Standard input	Standard output	Base price applies to
Qwen3.7 Max	阿里巴巴	¥12 / 1M tokens	¥36 / 1M tokens	—
Claude Opus 4.6	Anthropic	$5 / 1M tokens	$25 / 1M tokens	—
Kimi K2.6	Facebook AI研究实验室	$0.95 / 1M tokens	$4 / 1M tokens	—
GLM 5.1	智谱AI	$1.4 / 1M tokens	$4.4 / 1M tokens	—

Version History

How each version of the Qwen3.7 Max series stacks up on benchmark tests

Qwen3.7 MaxQwen3.6-Max-Preview Qwen3.5 Max Preview Qwen3-Max-Thinking

Benchmark categories:

12 benchmarks with comparable scores. Each model shows its best score; mode label is displayed below.· Click a row to view its trend chart.

Benchmark	Qwen3.7 MaxCurrent	Qwen3.6-Max-Preview	Qwen3-Max-Thinking
HLE Accuracy 综合评估	53.50Thinking Enabled ｜ Tools	50.20Thinking Enabled ｜ Tools	49.80Thinking Enabled ｜ Tools
MMLU Pro Accuracy 综合评估	89.60Thinking Level · High	88.50Thinking Level · High	85.70Thinking Enabled
GPQA Diamond Accuracy 科学与综合推理	92.40Thinking Level · High	90.40Thinking Level · High	87.40Thinking Enabled
LiveCodeBench Pass @K 编程与软件工程	91.60Thinking Level · High	87.10Thinking Level · High	85.90Thinking Enabled
SWE-bench Multilingual Accuracy 编程与软件工程	78.30Thinking Enabled ｜ Tools	73.80Thinking Enabled ｜ Tools	--
SWE-Bench Pro - Public Accuracy 编程与软件工程	60.60Thinking Enabled ｜ Tools	57.30Deep Thinking Mode ｜ Tools	--
SWE-bench Verified Accuracy 编程与软件工程	80.40Thinking Enabled ｜ Tools	78.80Thinking Enabled ｜ Tools	75.30Thinking Enabled
SimpleBench Score (AVG@5) 常识推理	70.40Standard Mode	63.00Standard Mode	--
IF Bench Accuracy 指令跟随	79.10Thinking Level · High	74.20Thinking Level · High	70.90Thinking Enabled ｜ Tools
Terminal Bench 2.0 Accuracy AI Agent - 工具使用	69.70Thinking Enabled ｜ Tools	65.40Deep Thinking Mode ｜ Tools	--
Context Arena Accuracy (8 needles, 4K-128K context) 文本向量检索	56.01Standard Mode	88.14Thinking Enabled	--
IMO-AnswerBench Accuracy 数学推理	90.00Thinking Level · High	83.80Thinking Level · High	83.90Thinking Enabled

Single-Benchmark Version Trend

Viewing: HLE · 综合评估

Benchmark

NormalNormal + ToolsThinkingThinking + ToolsDeepDeep + Tools

X-axis shows model and release date, Y-axis shows score; solid lines connect the same mode across versions, while dotted guides align modes within the same generation.

Standard API Pricing Across the Qwen3.7 Max Series

Shows standard text input and output pricing side by side for each model. If extended-context pricing exists, the chart keeps the base rate and explains the threshold below.

Source: DataLearnerAI. Standard text prices shown here use the default supplier.

These models use different currencies or billing units, so the page falls back to raw price values instead of a shared bar chart.

Qwen3.7 Max

Supplier: 阿里巴巴

Standard input: ¥12 / 1M tokens

Standard output: ¥36 / 1M tokens

Qwen3.6-Max-Preview

Supplier: 阿里巴巴

Standard input: $1.3 / 1M tokens

Standard output: $7.8 / 1M tokens

Base price applies to <= 128

Qwen3-Max-Thinking

Supplier: 阿里巴巴

Standard input: $1.2 / 1M tokens

Standard output: $6 / 1M tokens

Base price applies to <= 32000

Model	Supplier	Standard input	Standard output	Base price applies to
Qwen3.7 Max	阿里巴巴	¥12 / 1M tokens	¥36 / 1M tokens	—
Qwen3.6-Max-Preview	阿里巴巴	$1.3 / 1M tokens	$7.8 / 1M tokens	<= 128
Qwen3-Max-Thinking	阿里巴巴	$1.2 / 1M tokens	$6 / 1M tokens	<= 32000