跳转到主要内容
AITier

Qwen3.5 122B A10B (Non-reasoning)

AlibabaQwen开源权重Apache 2.0 · 商用许可

描述

Qwen3.5-122B-A10B is a multimodal Mixture-of-Experts model with 122 billion total parameters and 10 billion activated parameters. It combines strong reasoning, coding, long-context, and visual understanding performance with production-friendly efficiency and a native 262K context window.

发布日期
2026-02-24
参数规模
122.0B
上下文长度
262K
支持模态
audio, image, text, video

能力雷达图

26
general
42
coding
83
reasoning
53
science
60
agents
80
multimodal

排行榜排名

领域#排名分数来源
智能体能力模型榜79
45.0
LS
代码能力榜184
57.0
AA
通用能力榜193
57.0
AA
多模态榜24
59.0
LS
科学能力181
57.0
AA

基准测试分数 (LLM Stats)

(LLM Stats (zeroeval))

3d

SUNRGBD0.36 / 100自报
Hypersim0.13 / 100自报

Agents

t2-bench79.5%自报
BFCL-V472.2%自报
AndroidWorld_SR66.4%自报
BrowseCompOpenAI (2025)63.8%自报
FullStackBench en62.6%自报
WideSearch60.5%自报
FullStackBench zh58.7%自报
OSWorld-Verified58.0%自报
TIR-Bench53.2%自报
Terminal-Bench 2.0Stanford × Laude Institute (2026)49.4%自报
VITA-Bench33.6%自报
DeepPlanning24.1%自报

Biology

GPQANYU + Cohere + Anthropic (2023)86.6%自报

Chemistry

SuperGPQA67.1%自报

Code

SWE-Bench Verified72.0%自报

Communication

Multi-Challenge61.5%自报

Embodied

EmbSpatialBench0.84 / 100自报

Finance

MMLU-Pro86.7%自报
MMLU-ProX82.2%自报

General

MMLU-Redux94.0%自报
IFEvalGoogle Research (2023)93.4%自报
C-Eval91.9%自报
Global PIQA88.4%自报
MAXIFE87.9%自报
MMMLU86.7%自报
MMMU83.9%自报
MMStar82.9%自报
Include82.8%自报
LiveCodeBench v678.9%自报
MMMU-Pro76.9%自报
IFBench76.1%自报
SimpleVQA0.62 / 100自报
LongBench v260.2%自报
NOVA-6358.6%自报

Grounding

RefCOCO-avg0.91 / 100自报
ScreenSpot Pro70.4%自报
RefSpatialBench0.69 / 100自报

Healthcare

VideoMMMU82.0%自报
SlakeVQA81.6%自报
MedXpertQA67.3%自报
PMC-VQA63.3%自报

Image To Text

OCRBench92.1%自报

Language

LingoQA80.8%自报
WMT24++78.3%自报

Long Context

MLVU87.3%自报
LVBench74.4%自报
AA-LCR66.9%自报
MMLongBench-Doc0.59 / 100自报

Math

HMMT 202591.4%自报
HMMT2590.3%自报
MathVista-Mini87.4%自报
MathVision86.2%自报
DynaMath85.9%自报
CodeForces0.85 / 3000自报
PolyMATH68.9%自报
Humanity's Last Exam47.5%自报

Multimodal

VLMsAreBlind96.7%自报
AI2D93.3%自报
V*93.2%自报
MMBench-V1.192.8%自报
OmniDocBench 1.589.8%自报
VideoMME w sub.87.3%自报
VideoMME w/o sub.83.9%自报
CC-OCR81.8%自报
CharXiv-R77.2%自报
MVBench76.6%自报
MMVU74.7%自报
BabyVision40.2%自报
ZEROBench-Sub0.36 / 100自报
Nuscene15.4%自报
ZEROBench0.09 / 100自报

Reasoning

CountBench0.97 / 100自报
BrowseComp-zh69.9%自报
Hallusion Bench67.6%自报
ERQA62.0%自报
Seal-044.1%自报
OJBench39.5%自报

Spatial Reasoning

RealWorldQA85.1%自报

Vision

ODinW44.5%自报

AA 评测指数

(Artificial Analysis)
Coding Index(Artificial Analysis)
43.3
Intelligence Index(Artificial Analysis)
28.2
Tau2(Sierra + U Toronto + Vector Institute (2025))
0.8
Gpqa(NYU + Cohere + Anthropic (2023))
0.8
Lcr(Artificial Analysis)
0.6
Ifbench(Google Research (2023))
0.5
Terminalbench V2 1
0.5
Scicode(UIUC + Argonne National Lab (2024))
0.4
Terminalbench Hard(Stanford × Laude Institute (2026))
0.3
Hle(Center for AI Safety + Scale AI (2025))
0.2
Tau Banking
0.1

LLM Stats 分类评分

(LLM Stats (zeroeval))
Biology
90
Legal
80
Math
80
Physics
80
Structured Output
80
Image To Text
80
Instruction Following
80
Language
80
Embodied
80
Finance
80
Grounding
80
Healthcare
80
Chemistry
80
Text-to-image
80
Video
80
Long Context
70
Multimodal
70
Reasoning
70
Spatial Reasoning
70
Frontend Development
70
General
70
Economics
70
Vision
70
Search
60
Agents
60
Code
60
Communication
60
Tool Calling
60
Spatial
20
3d
20

定价

输入价格$0.4 / 1M tokens
输出价格$3.2 / 1M tokens
混合价格(3:1)$1.1 / 1M tokens

速度

Tokens/秒148.9
首Token延迟1.03s
首回答延迟1.03s

供应商价格排行

供应商价格排行

3 个供应商

最便宜: Alibaba最贵: Cortecs
供应商输入输出
1Alibaba主要
$0.4
$3.2
2NanoGPT
$0.437
$3.496
3Cortecs
$0.495
$3.46

比较该模型在不同 API 供应商之间的定价。

外部链接

LLM StatsArtificial Analysis