跳轉到主要內容

Qwen3 VL 30B A3B (Reasoning)

AlibabaQwen開源權重Apache 2.0 · 商用許可

描述

Qwen3-VL is a large multimodal model that unifies vision, language, and reasoning to achieve human-level perception and cognition across text, images, and video. Built on a 235B-parameter architecture, it integrates early joint training of visual and textual modalities for strong language grounding. The model supports up to a 1 million-token context window and excels at visual understanding, spatial reasoning, long video comprehension, and tool-based interaction. It can generate code from images, perform precise 2D/3D object grounding, and operate digital interfaces like a visual agent. The “Instruct” version rivals Gemini 2.5 Pro in perception benchmarks, while the “Thinking” version leads in multimodal reasoning and STEM tasks. With multilingual OCR, creative writing, and fine-grained scene interpretation, Qwen3-VL establishes a new open-source frontier for integrated vision-language intelligence.

發布日期
2025-10-03
參數規模
31.0B
上下文長度
131K
支援模態
image, text, video

能力雷達圖

33
general
61
coding
81
reasoning
44
science估算
70
agents
100
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69
54.0
LS
程式碼能力榜211
46.0
AA
通用能力榜276
39.0
AA
數學推理73
83.0
AA
多模態榜42
77.0
LS
推理能力85
53.0
LS
科學能力247
45.0
AA

基準測試分數 (LLM Stats)

3d

BLINK65.4%自報

Agents

BFCL-v368.6%自報
OSWorld30.6%自報

Biology

GPQA74.4%自報

Chemistry

SuperGPQA56.4%自報

Communication

MM-MT-Bench7.90 / 100自報
WritingBench85.2%自報
Multi-IF73.0%自報

Creativity

Creative Writing v382.5%自報
Arena-Hard v256.7%自報

Factuality

SimpleQA23.9%自報

Finance

MMLU87.6%自報
MMLU-Pro80.5%自報
MMLU-ProX76.1%自報

General

MMLU-Redux90.9%自報
IFEval81.7%自報
MLVU-M78.9%自報
MMMU (val)76.0%自報
MMStar75.5%自報
Include74.5%自報
LiveBench 2024112572.1%自報
LiveCodeBench v664.2%自報
MMMU-Pro63.0%自報

Grounding

ScreenSpot94.7%自報
ScreenSpot Pro57.3%自報

Healthcare

VideoMMMU75.0%自報

Image To Text

OCRBench83.9%自報
OCRBench-V2 (en)62.6%自報
OCRBench-V2 (zh)60.4%自報

Language

CharadesSTA62.7%自報

Long Context

LVBench59.2%自報

Math

AIME 202583.1%自報
MathVista-Mini81.9%自報
HMMT2567.6%自報
MathVision65.7%自報
PolyMATH51.7%自報

Multimodal

DocVQAtest95.0%自報
MMBench-V1.188.9%自報
CharXiv-D86.9%自報
AI2D86.9%自報
InfoVQAtest86.0%自報
CC-OCR77.8%自報
MuirBench77.6%自報
Video-MME73.3%自報
MVBench72.0%自報
CharXiv-R56.6%自報

Reasoning

Hallusion Bench66.0%自報
ERQA45.3%自報

Spatial Reasoning

RealWorldQA77.4%自報

Vision

ODinW42.3%自報

AA 評測指數

Math Index
82.3
Intelligence Index
13.3
Aime 25
0.8
Mmlu Pro
0.8
Gpqa
0.7
Livecodebench
0.7
Ifbench
0.5
Lcr
0.4
Scicode
0.3
Tau2
0.2
Hle
0.1
Terminalbench Hard
0.1

LLM Stats 分類評分

Communication
3
Multimodal
100
Instruction Following
80
Language
80
Legal
80
Structured Output
80
Finance
80
Grounding
80
Healthcare
80
Text-to-image
80
Image To Text
70
Math
70
Physics
70
Reasoning
70
Spatial Reasoning
70
General
70
3d
70
Biology
70
Chemistry
70
Creativity
70
Tool Calling
70
Video
70
Vision
70
Writing
70
Long Context
60
Economics
60
Agents
50
Factuality
20

定價

輸入價格$0.2 / 1M tokens
輸出價格$0.75 / 1M tokens
混合價格(3:1)$0.338 / 1M tokens

速度

Tokens/秒122.7
首Token延遲1.01s
首回答延遲17.31s

供應商價格排行

供應商價格排行

2 個供應商

最便宜: Alibaba (China)最貴: Alibaba
供應商輸入輸出
1Alibaba (China)最便宜
$0.108
$0.431
2Alibaba主要
$0.2
$0.75

比較該模型在不同 API 供應商之間的定價。

外部連結