Saltar al contenido principal

Qwen2.5 VL 72B Instruct

Alibaba Cloud / Qwen TeamQwenOpen Weighttongyi-qianwen

Descripción

Qwen2.5-VL is the new flagship vision-language model of Qwen, significantly improved from Qwen2-VL. It excels at recognizing objects, analyzing text/charts/layouts in images, acting as a visual agent, understanding long videos (over 1 hour) with event pinpointing, performing visual localization (bounding boxes/points), and generating structured outputs from documents.

Fecha de lanzamiento
2025-01-26
Parámetros
72.0B
Longitud del contexto
32K
Modalidades
image, text

Radar de capacidades

50
general
0
coding
60
reasoning
60
scienceest.
40
agents
80
multimodal

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

Rankings

Dominio#PosiciónPuntuaciónFuente
Agents & Tools82
45.0
LS
Multimodal Ranking54
73.0
LS
Reasoning74
55.0
LS

Puntuaciones de benchmarks (LLM Stats)

Agents

AITZ_EM83.2%Aut.
MobileMiniWob++_SR68.0%Aut.
AndroidWorld_SR35.0%Aut.
OSWorld8.8%Aut.

General

MMVet76.2%Aut.
MLVU-M74.6%Aut.
MMStar70.8%Aut.
MMMU70.2%Aut.
MMMU-Pro51.1%Aut.

Grounding

ScreenSpot87.1%Aut.
ScreenSpot Pro43.6%Aut.

Image To Text

DocVQA96.4%Aut.
OCRBench88.5%Aut.
OCRBench-V2 (en)61.5%Aut.

Long Context

EgoSchema76.2%Aut.
LVBench47.3%Aut.

Math

MathVista-Mini74.8%Aut.
MathVision38.1%Aut.

Multimodal

Android Control Low_EM93.7%Aut.
ChartQA89.5%Aut.
AI2D88.4%Aut.
MMBench88.0%Aut.
CC-OCR79.8%Aut.
TempCompass74.8%Aut.
VideoMME w/o sub.73.3%Aut.
PerceptionTest73.2%Aut.
MVBench70.4%Aut.
Android Control High_EM67.4%Aut.
MMBench-Video2.0%Aut.

Reasoning

Hallusion Bench55.2%Aut.

Índices de evaluación AA

No hay datos de evaluación AA disponibles

Puntuaciones por categoría LLM Stats

Structured Output
80
Text-to-image
80
Image To Text
80
Spatial Reasoning
70
Grounding
70
Healthcare
70
Reasoning
70
Vision
60
Long Context
60
Math
60
Multimodal
60
Video
50
General
50
Agents
40

Precios

Precio de entrada$0.25 / 1M tokens
Precio de salida$0.75 / 1M tokens
Precio mixto (3:1)$0.375 / 1M tokens

Velocidad

No hay datos de velocidad disponibles

Proveedores disponibles

(Unidades internas LS)

No hay datos de proveedores disponibles

Fuentes externas