メインコンテンツへスキップ

Qwen3 VL 4B (Reasoning)

AlibabaQwenオープンウエイトApache 2.0 · 商用利用可

説明

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.

リリース日
2025-10-14
パラメータ
4.0B
コンテキスト長
—
モダリティ
image, text

能力レーダー

26
general
32
coding
30
reasoning
36
science
70
agents
100
multimodal

ランキング

ドメイン#順位スコアソース
コーディングランキング506
20.0
AA
総合ランキング498
28.0
AA
マルチモーダルランキング140
25.0
LS
科学513
26.0
AA

ベンチマークスコア (LLM Stats)

(LLM Stats (zeroeval))

Chat

MM-MT-Bench7.70 / 100自己申告
IFEvalGoogle Research (2023)82.6%自己申告
Multi-IF73.6%自己申告

Creativity

Creative Writing v376.1%自己申告

General

MMLU81.5%自己申告
MLVU-M75.7%自己申告
BFCL-v367.3%自己申告
Include64.6%自己申告
Arena-Hard v236.8%自己申告

Language

MMLU-Redux86.0%自己申告
MMLU-Pro73.6%自己申告
MMLU-ProX65.0%自己申告

Math

MathVista-Mini79.5%自己申告
AIME 202574.5%自己申告
LiveBench 2024112568.4%自己申告
MathVision60.0%自己申告
HMMT2553.1%自己申告
PolyMATH44.6%自己申告

Multimodal

VideoMMMU69.4%自己申告
OSWorld31.4%自己申告

Reasoning

CharXiv-D83.9%自己申告
GPQANYU + Cohere + Anthropic (2023)64.1%自己申告
LiveCodeBench v651.3%自己申告
CharXiv-R50.3%自己申告
SuperGPQA46.8%自己申告

Video

CharadesSTA59.0%自己申告

Vision

DocVQAtest94.2%自己申告
ScreenSpot92.9%自己申告
MMBench-V1.186.7%自己申告
AI2D84.9%自己申告
InfoVQAtest83.0%自己申告
OCRBench80.8%自己申告
MuirBench75.0%自己申告
CC-OCR73.8%自己申告
MMStar73.2%自己申告
RealWorldQA73.2%自己申告
MMMU (val)70.8%自己申告
MVBench69.3%自己申告
Hallusion Bench64.1%自己申告
BLINK63.4%自己申告
OCRBench-V2 (en)61.8%自己申告
MMMU-Pro57.0%自己申告
OCRBench-V2 (zh)55.8%自己申告
LVBench53.5%自己申告
ScreenSpot Pro49.2%自己申告
ERQA47.3%自己申告
ODinW39.4%自己申告

Writing

WritingBench84.0%自己申告

AA評価指数

(Artificial Analysis)
Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))
70.0
Gpqa(NYU + Cohere + Anthropic (2023))
49.4
Ifbench(Google Research (2023))
36.6
Livecodebench(UC Berkeley + MIT + Cornell (2024))
32.0
Math Index(Artificial Analysis)
25.7
Aime 25(MAA (Mathematical Association of America))
25.7
Lcr(Artificial Analysis)
21.3
Tau2(Sierra + U Toronto + Vector Institute (2025))
15.5
Intelligence Index(Artificial Analysis)
7.0
Hle(Center for AI Safety + Scale AI (2025))
4.6
Terminalbench Hard(Stanford × Laude Institute (2026))
1.5

LLM Statsカテゴリスコア

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

価格設定

入力価格無料
出力価格無料
混合価格(3:1)無料

速度

トークン/秒0.0
初トークン遅延0.00s
初回答遅延0.00s

プロバイダー価格ランキング

プロバイダーデータがありません

外部リンク