跳轉到主要內容

Qwen3 235B A22B 2507 (Reasoning)

AlibabaQwen開源權重Apache 2.0 · 商用許可

描述

Qwen3-235B-A22B-Thinking-2507 is a state-of-the-art thinking-enabled Mixture-of-Experts (MoE) model with 235B total parameters (22B activated). It features 94 layers, 128 experts (8 activated), and supports 262K native context length. This version delivers significantly improved reasoning performance, achieving state-of-the-art results among open-source thinking models on logical reasoning, mathematics, science, coding, and academic benchmarks. Key enhancements include markedly better general capabilities (instruction following, tool usage, text generation), enhanced 256K long-context understanding, and increased thinking depth. The model supports only thinking mode with automatic <think> tag inclusion.

發布日期
2025-07-25
參數規模
235.0B
上下文長度
262K
支援模態
text

能力雷達圖

70
general
0
coding
70
reasoning
68
science估算
60
agents
0
multimodal

缺少專門科學評測時,Science 由 LLM Stats 科學得分或推理能力估算。

排行榜排名

暫無排名資料

基準測試分數 (LLM Stats)

(LLM Stats (zeroeval))

Chat

IFEvalGoogle Research (2023)87.8%自報
Multi-IF80.6%自報
TAU-bench Retail67.8%自報
Tau2 Airline58.0%自報

Code

CFEval2134.00 / 10000自報

Communication

Tau2 Retail71.9%自報
Tau2 Telecom45.6%自報

Creativity

Creative Writing v386.1%自報

General

Include81.0%自報
Arena-Hard v279.7%自報
BFCL-v371.9%自報

Language

MMLU-Redux93.8%自報
MMLU-Pro84.4%自報
MMLU-ProX81.0%自報

Math

AIME 202592.3%自報
HMMT2583.9%自報
LiveBench 2024112578.4%自報
PolyMATH60.1%自報

Reasoning

GPQANYU + Cohere + Anthropic (2023)81.1%自報
LiveCodeBench v674.1%自報
SuperGPQA64.9%自報
TAU-bench Airline46.0%自報
OJBench32.5%自報
Humanity's Last Exam18.2%自報

Writing

WritingBench88.3%自報

AA 評測指數

(Artificial Analysis)

暫無 AA 評測資料

LLM Stats 分類評分

(LLM Stats (zeroeval))
Instruction Following
80
Language
80
Legal
80
Structured Output
80
Finance
80
Healthcare
80
Biology
80
Creativity
80
Writing
80
Chat
70
Math
70
Physics
70
Reasoning
70
General
70
Agents
70
Chemistry
70
Communication
70
Multimodal
60
Spatial Reasoning
60
Economics
60
Tool Calling
60
Vision
40

定價

輸入價格$0.23 / 1M tokens
輸出價格$2.3 / 1M tokens
混合價格(3:1)$0.305 / 1M tokens
快取讀取價格$0.05 / 1M tokens

速度

Tokens/秒153.9
首Token延遲0.74s
首回答延遲13.74s

供應商價格排行

供應商價格排行

6 個供應商

最便宜: TensorX最貴: Jiekou.AI
供應商輸入輸出
1TensorX最便宜
$0.072
$0.464
2OpenRouter
$0.14
$0.8
3Kilo Gateway
$0.14
$0.8
4Alibaba主要
$0.23
$2.3
5NanoGPT
$0.3
$0.5
6Jiekou.AI
$0.3
$3

比較該模型在不同 API 供應商之間的定價。

外部連結