跳轉到主要內容
AITier

Qwen3.5 35B A3B (Non-reasoning)

AlibabaQwen開源權重Apache 2.0 · 商用許可

描述

Qwen3.5-35B-A3B is a multimodal Mixture-of-Experts model with 35 billion total parameters and 3 billion activated parameters. It combines strong reasoning, coding, agentic, and visual understanding performance with production-friendly efficiency and a native 262K context window.

發布日期
2026-02-24
參數規模
35.0B
上下文長度
262K
支援模態
audio, image, text, video

能力雷達圖

21
general
29
coding
82
reasoning
50
science估算
60
agents
80
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

領域#排名分數來源
智慧體能力模型榜53
56.0
LS
程式碼能力榜221
45.0
AA
通用能力榜183
52.0
AA
多模態榜70
69.0
LS
推理能力57
68.0
LS
科學能力167
52.0
AA

基準測試分數 (LLM Stats)

3d

SUNRGBD0.33 / 100自報
Hypersim0.13 / 100自報

Agents

t2-bench81.2%自報
AndroidWorld_SR71.1%自報
BFCL-V467.3%自報
BrowseComp61.0%自報
FullStackBench en58.1%自報
WideSearch57.1%自報
TIR-Bench55.5%自報
FullStackBench zh55.0%自報
OSWorld-Verified54.5%自報
Terminal-Bench 2.040.5%自報
VITA-Bench31.9%自報
DeepPlanning22.8%自報

Biology

GPQA84.2%自報

Chemistry

SuperGPQA63.4%自報

Code

SWE-Bench Verified69.2%自報

Communication

Multi-Challenge60.0%自報

Embodied

EmbSpatialBench0.83 / 100自報

Finance

MMLU-Pro85.3%自報
MMLU-ProX81.0%自報

General

MMLU-Redux93.3%自報
IFEval91.9%自報
C-Eval90.2%自報
MAXIFE86.6%自報
Global PIQA86.6%自報
MMMLU85.2%自報
MMStar81.9%自報
MMMU81.4%自報
Include79.7%自報
MMMU-Pro75.1%自報
LiveCodeBench v674.6%自報
IFBench70.2%自報
LongBench v259.0%自報
SimpleVQA0.58 / 100自報
NOVA-6357.1%自報

Grounding

RefCOCO-avg0.89 / 100自報
ScreenSpot Pro68.6%自報
RefSpatialBench0.64 / 100自報

Healthcare

VideoMMMU80.4%自報
SlakeVQA78.7%自報
PMC-VQA62.0%自報
MedXpertQA61.4%自報

Image To Text

OCRBench91.0%自報

Language

LingoQA79.2%自報
WMT24++76.3%自報

Long Context

MLVU85.6%自報
LVBench71.4%自報
MMLongBench-Doc0.59 / 100自報
AA-LCR58.5%自報

Math

HMMT2589.2%自報
HMMT 202589.0%自報
MathVista-Mini86.2%自報
DynaMath85.0%自報
MathVision83.9%自報
CodeForces0.82 / 3000自報
PolyMATH64.4%自報
Humanity's Last Exam47.4%自報

Multimodal

VLMsAreBlind97.0%自報
V*92.7%自報
AI2D92.6%自報
MMBench-V1.191.5%自報
OmniDocBench 1.589.3%自報
VideoMME w sub.86.6%自報
VideoMME w/o sub.82.5%自報
CC-OCR80.7%自報
CharXiv-R77.5%自報
MVBench74.8%自報
MMVU72.3%自報
BabyVision38.4%自報
ZEROBench-Sub0.34 / 100自報
Nuscene14.6%自報
ZEROBench0.08 / 100自報

Reasoning

CountBench0.98 / 100自報
BrowseComp-zh69.5%自報
Hallusion Bench67.9%自報
ERQA64.8%自報
Seal-041.4%自報
OJBench36.0%自報

Spatial Reasoning

RealWorldQA84.1%自報

Vision

ODinW42.6%自報

AA 評測指數

Intelligence Index
23.4
Tau2
0.9
Gpqa
0.8
Lcr
0.6
Ifbench
0.4
Scicode
0.3
Hle
0.1
Terminalbench Hard
0.1

LLM Stats 分類評分

Image To Text
80
Instruction Following
80
Language
80
Legal
80
Math
80
Physics
80
Structured Output
80
Embodied
80
Finance
80
General
80
Biology
80
Text-to-image
80
Video
80
Long Context
70
Multimodal
70
Reasoning
70
Spatial Reasoning
70
Frontend Development
70
Grounding
70
Healthcare
70
Chemistry
70
Vision
70
Search
60
Code
60
Communication
60
Economics
60
Tool Calling
60
Agents
50
3d
20
Spatial
10

定價

輸入價格$0.25 / 1M tokens
輸出價格$2 / 1M tokens
混合價格(3:1)$0.688 / 1M tokens

速度

Tokens/秒203.7
首Token延遲1.26s
首回答延遲1.26s

供應商價格排行

供應商價格排行

12 個供應商

最便宜: Novita最貴: Venice AI
供應商輸入輸出
1Novita最便宜
$0
$0
2OrcaRouter
$0.057
$0.459
3OpenRouter
$0.14
$1
4Deep Infra
$0.14
$1
5Kilo Gateway
$0.1625
$1.3
6NanoGPT
$0.225
$1.8
7SiliconFlow (China)
$0.23
$1.86
8Alibaba主要
$0.25
$2
9Hugging Face
$0.25
$2
10NovitaAI
$0.25
$2
11Mixlayer
$0.25
$1.3
12Venice AI
$0.3125
$1.25

比較該模型在不同 API 供應商之間的定價。

外部連結

LLM StatsArtificial Analysis