Qwen3 Max Thinking

AlibabaQwenProprietary

描述

The latest flagship reasoning model in the Qwen3 family. Further enhanced by multiple innovations like adaptive tool-use and advanced test-time scaling techniques

發布日期

2026-01-26

參數規模

1.0T

上下文長度

262K

支援模態

text

能力雷達圖

general

coding

reasoning

science

agents

multimodal

排行榜排名

領域	#排名	分數	來源
智慧體能力模型榜	79	41.0	LS
程式碼能力榜	238	61.0	AA
通用能力榜	158	59.0	AA
科學能力	147	67.0	AA

基準測試分數 (LLM Stats)

(LLM Stats (zeroeval))

Agents

VITA-Bench

40.9%自報

MCP-Mark

33.5%自報

DeepPlanning

28.7%自報

Toolathlon

18.8%自報

Chat

IFEvalGoogle Research (2023)

93.4%自報

Multi-Challenge

63.3%自報

Code

SecCodeBench

57.5%自報

General

C-Eval

93.7%自報

MAXIFE

84.0%自報

NOVA-63

54.2%自報

Instruction Following

IFBench

70.9%自報

Knowledge

HLE-Verified

37.6%自報

Language

MMLU-Redux

92.8%自報

MMLU-Pro

85.7%自報

MMMLU

84.4%自報

WMT24++

77.6%自報

Long Context

LongBench v2

60.6%自報

Math

AIME 2026

93.3%自報

IMO-AnswerBench

83.9%自報

Reasoning

GPQANYU + Cohere + Anthropic (2023)

87.4%自報

Global PIQA

86.0%自報

LiveCodeBench v6

85.9%自報

SWE-Bench Verified

75.3%自報

AA-LCR

68.7%自報

SuperGPQA

67.3%自報

SWE-bench Multilingual

66.7%自報

BrowseComp-zh

60.9%自報

BrowseCompOpenAI (2025)

53.9%自報

Humanity's Last Exam (with tools, text-only)

49.8%自報

Seal-0

46.9%自報

Humanity's Last Exam (no tools, text-only)

30.2%自報

Terminal-Bench 2.0Stanford × Laude Institute (2026)

22.5%自報

Search

WideSearch

57.9%自報

Tool Calling

BFCL-V4

67.7%自報

AA 評測指數

(Artificial Analysis)

Gpqa(NYU + Cohere + Anthropic (2023))

86.1

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

83.6

Lcr(Artificial Analysis)

74.3

Ifbench(Google Research (2023))

70.7

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

28.0

Terminalbench Hard(Stanford × Laude Institute (2026))

24.2

Intelligence Index(Artificial Analysis)

21.3

LLM Stats 分類評分

(LLM Stats (zeroeval))

Language

Biology

Chat

Instruction Following

Legal

Physics

Structured Output

Finance

Frontend Development

Healthcare

Chemistry

Math

Economics

Long Context

Reasoning

General

Communication

Code

Knowledge

Agents

Tool Calling

定價

輸入價格免費

輸出價格免費

混合價格(3:1)免費

速度

Tokens/秒0.0

首Token延遲0.00s

首回答延遲0.00s

供應商價格排行

1 個供應商

供應商輸入輸出

1DeepInfra

$0.00001

比較該模型在不同 API 供應商之間的定價。

外部連結

LLM Stats Artificial Analysis