Qwen3 235B A22B (Non-reasoning)
AlibabaQwenOpen WeightApache 2.0 · Commercial OK
描述
Qwen3 235B A22B is a large language model developed by Alibaba, featuring a Mixture-of-Experts (MoE) architecture with 235 billion total parameters and 22 billion activated parameters. It achieves competitive results in benchmark evaluations of coding, math, general capabilities, and more, compared to other top-tier models.
發布日期
2025-04-28
參數規模
235.0B
上下文長度
131K
支援模態
text
能力雷達圖
33
general
23
coding
40
reasoning
39
science估算
70
agents
0
multimodal
Science 在缺少專門科學評測時使用推理能力代理估算。
排行榜排名
基準測試分數 (LLM Stats)
Biology
GPQA
47.5%自報
Chemistry
SuperGPQA
44.1%自報
Code
EvalPlus
0.78 / 100自報
LiveCodeBench
70.7%自報
Aider
61.8%自報
Creativity
Arena Hard
95.6%自報
Finance
MMLU
87.8%自報
MMLU-Pro
68.2%自報
General
MMLU-Redux
87.4%自報
MMMLU
86.7%自報
MBPP
0.81 / 100自報
LiveBench
77.1%自報
Include
73.5%自報
MultiLF
71.9%自報
BFCL
70.8%自報
MultiPL-E
65.9%自報
Language
BBH
88.9%自報
Math
GSM8k
94.4%自報
AIME 2024
85.7%自報
MGSM
83.5%自報
AIME 2025
81.5%自報
MATH
71.8%自報
Reasoning
CRUX-O
0.79 / 100自報
AA 評測指數
Math Index23.7
Intelligence Index17.0
Coding Index14.0
Math 5000.9
Mmlu Pro0.8
Gpqa0.6
Ifbench0.4
Livecodebench0.3
Aime0.3
Scicode0.3
Tau20.3
Aime 250.2
Terminalbench Hard0.1
Hle0.0
Lcr0.0
LLM Stats 分類評分
Writing100
Creativity100
Language80
Math80
Reasoning80
Tool Calling70
Code70
Finance70
General70
Healthcare70
Legal70
Biology50
Chemistry50
Physics50
Economics40
定價
輸入價格$0.45 / 1M tokens
輸出價格$1.8 / 1M tokens
混合價格(3:1)$0.787 / 1M tokens
速度
Tokens/秒64.1 tokens/s
首Token延遲1.24s
首回答延遲1.24s
可用提供商
(LS 內部計價單位)暫無提供商資料