Qwen3.7 Max

AlibabaQwenProprietary

描述

Qwen3.7 Max is Alibaba Cloud Qwen Team's proprietary flagship model for agent-driven workflows. It is designed for coding agents, office automation, MCP and multi-agent orchestration, and long-horizon autonomous execution, with a 1 million token context window and up to 65,536 output tokens. Qwen reports strong agentic coding results including 69.7 on Terminal-Bench 2.0-Terminus, 80.4 on SWE-bench Verified, 60.6 on SWE-Pro, and 78.3 on SWE-Multilingual, alongside 92.4 on GPQA Diamond and 97.1 on HMMT 2026 Feb.

发布日期

2026-05-19

参数规模

—

上下文长度

1.0M

支持模态

text

能力雷达图

general

coding

reasoning

science

agents

multimodal

排行榜排名

领域	#排名	分数	来源
智能体能力模型榜	7	65.0	LS
代码能力榜	55	84.0	AA
通用能力榜	31	85.0	AA
数学推理	30	85.0	LB
推理能力	26	83.0	LB
科学能力	48	83.0	AA

基准测试分数 (LLM Stats)

(LLM Stats (zeroeval))

Agents

QwenSVG

1608.00 / 2000自报

Kernel Bench L3

96.0%自报

SpreadSheetBench-v1

87.0%自报

CoWorkBench

67.2%自报

MCP-Mark

60.8%自报

QwenWorldBench

57.3%自报

Finance Agent v2

48.4%

VITA-Bench

47.9%自报

Chat

IFEvalGoogle Research (2023)

94.3%自报

Code

QwenWebBench

1568.00 / 2000自报

Claw-Eval

65.2%自报

ZClawBench

64.3%自报

SkillsBench

59.2%自报

NL2Repo

47.2%自报

General

MAXIFE

89.2%自报

Include

86.2%自报

NOVA-63

59.0%自报

Instruction Following

IFBench

79.1%自报

Language

MMLU-Redux

95.0%自报

MMMLU

90.3%自报

MMLU-Pro

89.6%自报

MMLU-ProX

87.0%自报

WMT24++

85.8%自报

Long Context

MRCR 128K (8-needle)

90.4%自报

Math

HMMT Feb 26

97.1%自报

IMO-AnswerBench

90.0%自报

PolyMATH

86.5%自报

LiveBench

74.3%

MathArena Apex

44.5%自报

Reasoning

GPQANYU + Cohere + Anthropic (2023)

92.4%自报

LiveCodeBench v6

91.6%自报

Global PIQA

91.4%自报

SWE-Bench Verified

80.4%自报

SWE-bench Multilingual

78.3%自报

MCP Atlas

76.4%自报

SuperGPQA

73.6%自报

Terminal-Bench 2.0Stanford × Laude Institute (2026)

69.7%自报

SWE-Bench ProPrinceton NLP (2024)

60.6%自报

SciCode

53.5%自报

Humanity's Last Exam

41.4%自报

CritPT

11.4%自报

Tool Calling

BFCL-V4

75.0%自报

AA 评测指数

(Artificial Analysis)

Tau2(Sierra + U Toronto + Vector Institute (2025))

94.7

Gpqa(NYU + Cohere + Anthropic (2023))

92.3

Ifbench(Google Research (2023))

80.5

Lcr(Artificial Analysis)

74.7

Terminalbench V2 1

74.5

Coding Index(Artificial Analysis)

66.0

Terminalbench Hard(Stanford × Laude Institute (2026))

50.8

Scicode(UIUC + Argonne National Lab (2024))

48.8

Intelligence Index(Artificial Analysis)

46.7

Hle(Center for AI Safety + Scale AI (2025))

40.5

Tau Banking

11.8

LLM Stats 分类评分

(LLM Stats (zeroeval))

Chat

Instruction Following

Language

Multimodal

Spatial Reasoning

Structured Output

Legal

Physics

Productivity

Frontend Development

Healthcare

Math

Reasoning

Finance

General

Biology

Chemistry

Code

Economics

Tool Calling

Agents

Vision

定价

输入价格$2.5 / 1M tokens

输出价格$7.5 / 1M tokens

混合价格(3:1)$3.75 / 1M tokens

缓存读取价格$0.5 / 1M tokens

缓存写入价格$3.125 / 1M tokens

速度

Tokens/秒0.0

首Token延迟0.00s

首回答延迟0.00s

供应商价格排行

24 个供应商

最便宜: Novita最贵: Modelis

供应商输入输出

1Novita最便宜

2Together

$0.00001

3Merge Gateway

$0.825

$2.4755

4NovitaAI

$1.25

$3.75

5Kilo Gateway

$1.25

$3.75

6DevPass (LLM Gateway)

$1.25

$3.75

7OrcaRouter

$1.25

$3.75

8Pioneer

$1.25

$3.75

9OpenRouter

$1.475

$4.425

10AIHubMix

$1.69

$5.07

11CrossModel

$1.88

$5.63

12Alibaba主要

$2.5

$7.5

13NanoGPT

$2.5

$7.5

14Abacus

$2.5

$7.5

15OpenCode Go

$2.5

$7.5

16Alibaba (China)

$2.5

$7.5

17ZenMux

$2.5

$7.5

18Alibaba Coding Plan

$2.5

$7.5

19Requesty

$2.5

$7.5

20Alibaba Coding Plan (China)

$2.5

$7.5

21EmpirioLabs AI

$2.5

$7.5

22Charm Hyper

$2.5

$7.5

23Impossibl

$2.5

$7.5

24Modelis

比较该模型在不同 API 供应商之间的定价。

外部链接

LLM Stats Artificial Analysis