Saltar al contenido principal

Qwen2.5 VL 7B Instruct

Alibaba Cloud / Qwen TeamQwenOpen WeightApache 2.0 · Commercial OK

Descripción

Qwen2.5-VL is a vision-language model from the Qwen family. Key enhancements include visual understanding (objects, text, charts, layouts), visual agent capabilities (tool use, computer/phone control), long video comprehension with event pinpointing, visual localization (bounding boxes/points), and structured output generation.

Fecha de lanzamiento
2025-01-26
Parámetros
8.3B
Longitud del contexto
Modalidades

Radar de capacidades

50
general
0
coding
50
reasoning
51
scienceest.
50
agents
90
multimodal

Science usa un proxy de razonamiento cuando los benchmarks científicos dedicados no están disponibles.

Rankings

Dominio#PosiciónPuntuaciónFuente
Agents & Tools35
62.0
LS
Multimodal Ranking59
71.0
LS
Reasoning82
53.0
LS

Puntuaciones de benchmarks (LLM Stats)

Agents

MobileMiniWob++_SR91.4%Aut.
AITZ_EM81.9%Aut.
AndroidWorld_SR25.5%Aut.

General

MMVet67.1%Aut.
MMStar63.9%Aut.
MMT-Bench63.6%Aut.
MMMU58.6%Aut.
MMMU-Pro38.3%Aut.

Grounding

ScreenSpot84.7%Aut.
ScreenSpot Pro29.0%Aut.

Image To Text

DocVQA95.7%Aut.
OCRBench86.4%Aut.
TextVQA84.9%Aut.

Language

CharadesSTA43.6%Aut.

Long Context

MLVU70.2%Aut.
LongVideoBench54.7%Aut.
LVBench45.3%Aut.

Math

MathVista-Mini68.2%Aut.
MathVision25.1%Aut.

Multimodal

Android Control Low_EM91.4%Aut.
ChartQA87.3%Aut.
MMBench84.3%Aut.
InfoVQA82.6%Aut.
CC-OCR77.8%Aut.
TempCompass71.7%Aut.
VideoMME w sub.71.6%Aut.
PerceptionTest70.5%Aut.
MVBench69.6%Aut.
VideoMME w/o sub.65.1%Aut.
Android Control High_EM60.1%Aut.
MMBench-Video1.8%Aut.

Reasoning

Hallusion Bench52.9%Aut.

Índices de evaluación AA

No hay datos de evaluación AA disponibles

Puntuaciones por categoría LLM Stats

Image To Text
90
Structured Output
80
Text-to-image
80
Spatial Reasoning
60
Vision
60
Grounding
60
Healthcare
60
Long Context
60
Multimodal
60
Reasoning
60
Video
50
Agents
50
General
50
Math
50
Language
40

Precios

No hay datos de precios disponibles

Velocidad

No hay datos de velocidad disponibles

Proveedores disponibles

(Unidades internas LS)

No hay datos de proveedores disponibles

Fuentes externas