Saltar al contenido principal

Qwen3 VL 30B A3B Instruct

AlibabaQwenOpen WeightApache 2.0 · Uso Comercial

Descripción

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.

Fecha de lanzamiento
2025-10-03
Parámetros
31.0B
Longitud del contexto
262K
Modalidades
image, text, video

Radar de capacidades

29
general
44
coding
72
reasoning
43
science
70
agents
100
multimodal

Rankings

Dominio#PosiciónPuntuaciónFuente
Capacidad agéntica134
33.0
LS
Ranking de codificación347
31.0
AA
Ranking general409
32.0
AA
Ranking multimodal71
36.0
LS
Ciencia316
44.0
AA

Puntuaciones de benchmarks (LLM Stats)

(LLM Stats (zeroeval))

3d

BLINK67.7%Aut.

Agents

BFCL-v366.3%Aut.
OSWorld30.3%Aut.

Biology

GPQANYU + Cohere + Anthropic (2023)70.4%Aut.

Chemistry

SuperGPQA53.1%Aut.

Communication

MM-MT-Bench8.10 / 100Aut.
WritingBench82.6%Aut.
Multi-IF66.1%Aut.

Creativity

Creative Writing v384.6%Aut.
Arena-Hard v258.5%Aut.

Factuality

SimpleQA27.0%Aut.

Finance

MMLU85.0%Aut.
MMLU-Pro77.8%Aut.
MMLU-ProX70.9%Aut.

General

MMLU-Redux88.4%Aut.
IFEvalGoogle Research (2023)85.8%Aut.
MLVU-M81.3%Aut.
MMMU (val)74.2%Aut.
MMStar72.1%Aut.
Include71.6%Aut.
LiveBench 2024112565.4%Aut.
MMMU-Pro60.4%Aut.
LiveCodeBench v642.6%Aut.

Grounding

ScreenSpot94.7%Aut.
ScreenSpot Pro60.5%Aut.

Healthcare

VideoMMMU68.7%Aut.

Image To Text

OCRBench90.3%Aut.
OCRBench-V2 (en)63.2%Aut.
OCRBench-V2 (zh)57.8%Aut.

Language

CharadesSTA63.5%Aut.

Long Context

LVBench62.5%Aut.

Math

MathVista-Mini80.1%Aut.
AIME 202569.3%Aut.
MathVision60.2%Aut.
HMMT2550.6%Aut.
PolyMATH44.3%Aut.

Multimodal

DocVQAtest95.0%Aut.
MMBench-V1.187.0%Aut.
CharXiv-D85.5%Aut.
AI2D85.0%Aut.
InfoVQAtest82.0%Aut.
CC-OCR80.7%Aut.
Video-MME74.5%Aut.
MVBench72.3%Aut.
MuirBench62.9%Aut.
CharXiv-R48.9%Aut.

Reasoning

Hallusion Bench61.5%Aut.
ERQA43.0%Aut.

Spatial Reasoning

RealWorldQA73.7%Aut.

Vision

ODinW47.5%Aut.

Índices de evaluación AA

(Artificial Analysis)
Math Index(Artificial Analysis)
72.3
Intelligence Index(Artificial Analysis)
9.9
Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))
0.8
Aime 25(MAA (Mathematical Association of America))
0.7
Gpqa(NYU + Cohere + Anthropic (2023))
0.7
Livecodebench(UC Berkeley + MIT + Cornell (2024))
0.5
Ifbench(Google Research (2023))
0.3
Scicode(UIUC + Argonne National Lab (2024))
0.3
Lcr(Artificial Analysis)
0.3
Tau2(Sierra + U Toronto + Vector Institute (2025))
0.2
Hle(Center for AI Safety + Scale AI (2025))
0.1
Terminalbench Hard(Stanford × Laude Institute (2026))
0.1

Puntuaciones por categoría LLM Stats

(LLM Stats (zeroeval))
Communication
3
Multimodal
100
Structured Output
80
Instruction Following
80
Language
80
General
80
Grounding
80
Creativity
80
Text-to-image
80
Writing
80
Legal
70
Math
70
Reasoning
70
Spatial Reasoning
70
Image To Text
70
Finance
70
Healthcare
70
3d
70
Biology
70
Tool Calling
70
Video
70
Vision
70
Long Context
60
Physics
60
Chemistry
60
Agents
50
Economics
50
Factuality
30

Precios

Precio de entrada$0.2 / 1M tokens
Precio de salida$0.8 / 1M tokens
Precio mixto (3:1)$0.35 / 1M tokens

Velocidad

Tokens/seg0.0
Retraso del primer token0.00s
Tiempo hasta la respuesta0.00s

Ranking de Precios por Proveedor

Ranking de Precios por Proveedor

5 proveedores

Más barato: Kilo GatewayMás caro: NovitaAI
ProveedorEntradaSalida
1Kilo GatewayMás barato
$0.13
$0.52
2OpenRouter
$0.15
$0.6
3LLM Gateway
$0.15
$0.6
4AlibabaPRINCIPAL
$0.2
$0.8
5NovitaAI
$0.2
$0.7

Comparar precios entre diferentes proveedores de API para este modelo.

Fuentes externas