Qwen3.8-Max Benchmark Details

Qwen3.8-Max currently shows benchmark results led by Context Arena (3 / 126, score 96.92), LongBench v2 (1 / 13, score 66.30), HLE (15 / 189, score 56.20). 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

Qwen3.8-Max

Benchmark Results

General Knowledge

2 evaluations

Benchmark / mode

Score

Rank/total

HLE

Extra-High

43.60

54 / 189

HLE

Extra-HighTools

56.20

15 / 189

Other

1 evaluations

Benchmark / mode

Score

Rank/total

GPQA Diamond

Extra-High

92.60

24 / 270

Writing and Creative Capabilities

1 evaluations

Benchmark / mode

Score

Rank/total

Creative Writing

Standard Mode

1842

10 / 99

Common Sense Reasoning

1 evaluations

Benchmark / mode

Score

Rank/total

SimpleBench

Extra-High

62.50

22 / 92

Coding and Software Engineer

5 evaluations

Benchmark / mode

Score

Rank/total

FrontierSWE

Extra-HighTools

73.50

4 / 4

SWE-Bench Pro - Public

Extra-HighTools

67.70

5 / 60

DeepSWE

Extra-HighTools

56.60

21 / 34

NL2Repo-Bench

Extra-HighTools

55.90

7 / 12

MLS Bench

Extra-HighTools

3 / 5

Text Embedding

3 evaluations

Benchmark / mode

Score

Rank/total

Context Arena

Low

92.79

9 / 126

Context Arena

Thinking Level · Medium

95.15

5 / 126

Context Arena

Extra-High

96.92

3 / 126

AI Agent - Tool Usage

2 evaluations

Benchmark / mode

Score

Rank/total

Terminal-Bench 2.1

Extra-HighTools

86.60

10 / 49

Toolathlon-Verified

Extra-HighTools

72.50

9 / 10

Long Context

1 evaluations

Benchmark / mode

Score

Rank/total

LongBench v2

Extra-High

66.30

1 / 13

Agent Level Benchmark

1 evaluations

Benchmark / mode

Score

Rank/total

Agents' Last Exam

Extra-HighTools

7 / 14

Productivity Knowledge

1 evaluations

Benchmark / mode

Score

Rank/total

AutomationBench

Extra-HighTools

27.30

12 / 16

Math and Reasoning

2 evaluations

Benchmark / mode

Score

Rank/total

FrontierMath v2

Extra-High

74.74

11 / 58

FrontierMath Tier 4 v2

Extra-High

46.34

11 / 40

Compare with other models

Competitor Comparison

Benchmark scores for Qwen3.8-Max compared against top models in its class

Qwen3.8-MaxClaude Opus 5 GPT-5.6 Sol Kimi K3

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	Qwen3.8-MaxCurrent	Claude Opus 5	GPT-5.6 Sol	Kimi K3
HLE Accuracy 综合评估	56.20Thinking Level · Extra High ｜ Tools	64.70Thinking Level · High ｜ Tools	49.50Thinking Level · High	56.00Thinking Level · High ｜ Tools
GPQA Diamond Accuracy 科学与综合推理	92.60Thinking Level · Extra High	93.88Thinking Level · High	93.50Thinking Level · High	93.50Thinking Level · High
Creative Writing Elo、大模型评判两两对战写作和创作	1842.00Standard Mode	2116.10Standard Mode	1964.10Standard Mode	2070.80Standard Mode
SimpleBench Score (AVG@5) 常识推理	62.50Thinking Level · Extra High	80.60Thinking Level · High	64.80Thinking Level · Extra High	60.70Thinking Level · High
DeepSWE Pass@1 (DeepSWE v1.1) 编程与软件工程	56.60Thinking Level · Extra High ｜ Tools	68.80Thinking Level · High ｜ Tools	72.70Thinking Level · Extra High ｜ Tools	67.50Thinking Level · High ｜ Tools
FrontierSWE Dominance score 编程与软件工程	73.50Thinking Level · Extra High ｜ Tools	--	--	81.20Thinking Level · High ｜ Tools
MLS Bench Score 编程与软件工程	41.00Thinking Level · Extra High ｜ Tools	--	--	48.30Thinking Level · High ｜ Tools
SWE-Bench Pro - Public Accuracy 编程与软件工程	67.70Thinking Level · Extra High ｜ Tools	79.20Thinking Level · High ｜ Tools	64.60Thinking Level · Extra High ｜ Tools	--
Context Arena Accuracy (8 needles, 4K-128K context) 文本向量检索	96.92Thinking Level · Extra High	97.72Thinking Level · High	97.63Thinking Level · High	71.75Thinking Level · High
Terminal-Bench 2.1 Accuracy AI Agent - 工具使用	86.60Thinking Level · Extra High ｜ Tools	--	88.80Thinking Level · High	88.30Thinking Level · High ｜ Tools
Toolathlon-Verified Score AI Agent - 工具使用	72.50Thinking Level · Extra High ｜ Tools	--	--	76.50Thinking Level · High ｜ Tools
Agents' Last Exam Score Agent能力评测	27.00Thinking Level · Extra High ｜ Tools	--	52.70Thinking Level · Extra High ｜ Tools	28.30Thinking Level · High ｜ Tools

3 additional benchmarks remain in the chart above.

Standard API Pricing: Qwen3.8-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.

Qwen3.8-Max

Supplier: 阿里巴巴

Standard input: ¥12 / 1M tokens

Standard output: ¥36 / 1M tokens

Claude Opus 5

Supplier: Anthropic

Standard input: $5 / 1M tokens

Standard output: $25 / 1M tokens

GPT-5.6 Sol

Supplier: OpenAI

Standard input: $4 / 1M tokens

Standard output: $20 / 1M tokens

Kimi K3

Supplier: Moonshot AI

Standard input: ¥20 / 1M tokens

Standard output: ¥100 / 1M tokens

Model	Supplier	Standard input	Standard output	Base price applies to
Qwen3.8-Max	阿里巴巴	¥12 / 1M tokens	¥36 / 1M tokens	—
Claude Opus 5	Anthropic	$5 / 1M tokens	$25 / 1M tokens	—
GPT-5.6 Sol	OpenAI	$4 / 1M tokens	$20 / 1M tokens	—
Kimi K3	Moonshot AI	¥20 / 1M tokens	¥100 / 1M tokens	—

Version History

How each version of the Qwen3.8-Max series stacks up on benchmark tests

Qwen3.8-MaxQwen3.8-Max-Preview Qwen3.7 Max Qwen3.6-Max-Preview

Benchmark categories:

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

Benchmark	Qwen3.8-MaxCurrent	Qwen3.7 Max	Qwen3.6-Max-Preview
HLE Accuracy 综合评估	56.20Thinking Level · Extra High ｜ Tools	53.50Thinking Enabled ｜ Tools	50.20Thinking Enabled ｜ Tools
GPQA Diamond Accuracy 科学与综合推理	92.60Thinking Level · Extra High	92.40Thinking Level · High	90.40Thinking Level · High
SimpleBench Score (AVG@5) 常识推理	62.50Thinking Level · Extra High	70.40Standard Mode	63.00Standard Mode
SWE-Bench Pro - Public Accuracy 编程与软件工程	67.70Thinking Level · Extra High ｜ Tools	60.60Thinking Enabled ｜ Tools	57.30Deep Thinking Mode ｜ Tools
Context Arena Accuracy (8 needles, 4K-128K context) 文本向量检索	96.92Thinking Level · Extra High	56.01Standard Mode	88.14Thinking 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 Qwen3.8-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.

Qwen3.8-Max

Supplier: 阿里巴巴

Standard input: ¥12 / 1M tokens

Standard output: ¥36 / 1M tokens

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

Model	Supplier	Standard input	Standard output	Base price applies to
Qwen3.8-Max	阿里巴巴	¥12 / 1M tokens	¥36 / 1M tokens	—
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