Saltar al contenido principal

Qwen3 VL 8B 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
2025-10-14
Parámetros
9.0B
Longitud del contexto
262K
Modalidades
image, text, video

Radar de capacidades

80
general
0
coding
60
reasoning
34
scienceest.
70
agents
100
multimodal

Science se estima a partir de las puntuaciones científicas de LLM Stats o del razonamiento cuando no hay benchmarks científicos dedicados.

Rankings

Dominio#PosiciónPuntuaciónFuente
Ranking multimodal126
34.0
LS

Puntuaciones de benchmarks (LLM Stats)

(LLM Stats (zeroeval))

Chat

MM-MT-Bench7.70 / 100Aut.
IFEvalGoogle Research (2023)83.7%Aut.
Multi-IF75.1%Aut.

General

MMLU80.7%Aut.
MLVU-M78.1%Aut.
Include67.0%Aut.
BFCL-v366.3%Aut.

Language

MMLU-Redux84.9%Aut.
MMLU-Pro71.6%Aut.
MMLU-ProX65.4%Aut.

Math

MathVista-Mini77.2%Aut.
LiveBench 2024112562.0%Aut.
MathVision53.9%Aut.
AIME 202545.9%Aut.
HMMT2532.5%Aut.
PolyMATH30.4%Aut.

Multimodal

Video-MME71.4%Aut.
VideoMMMU65.3%Aut.
OSWorld33.9%Aut.

Reasoning

CharXiv-D83.0%Aut.
CharXiv-R46.4%Aut.
SuperGPQA44.5%Aut.
LiveCodeBench v639.3%Aut.

Video

CharadesSTA56.0%Aut.

Vision

DocVQAtest96.1%Aut.
ScreenSpot94.4%Aut.
OCRBench89.6%Aut.
AI2D85.7%Aut.
MMBench-V1.185.0%Aut.
InfoVQAtest83.1%Aut.
CC-OCR79.9%Aut.
RealWorldQA71.5%Aut.
MMStar70.9%Aut.
MMMU (val)69.6%Aut.
BLINK69.1%Aut.
MVBench68.7%Aut.
OCRBench-V2 (en)65.4%Aut.
MuirBench64.4%Aut.
OCRBench-V2 (zh)61.2%Aut.
Hallusion Bench61.1%Aut.
LVBench58.0%Aut.
MMMU-Pro55.9%Aut.
ScreenSpot Pro54.6%Aut.
ERQA45.8%Aut.
ODinW44.7%Aut.

Writing

WritingBench83.1%Aut.

Índices de evaluación AA

(Artificial Analysis)

No hay datos de evaluación AA disponibles

Puntuaciones por categoría LLM Stats

(LLM Stats (zeroeval))
Chat
3
Communication
3
Multimodal
100
Instruction Following
80
Structured Output
80
General
80
Creativity
80
Text-to-image
80
Writing
80
Image To Text
70
Language
70
Legal
70
Finance
70
Grounding
70
Healthcare
70
3d
70
Tool Calling
70
Vision
70
Long Context
60
Math
60
Reasoning
60
Spatial Reasoning
60
Video
60
Agents
50
Physics
40
Chemistry
40
Economics
40

Precios

Precio de entrada$0.25 / 1M tokens
Precio de salida$0.75 / 1M tokens
Precio mixto (3:1)$0.375 / 1M tokens
Precio de lectura caché$0.12 / 1M tokens

Velocidad

No hay datos de velocidad disponibles

Ranking de Precios por Proveedor

Ranking de Precios por Proveedor

1 proveedores

ProveedorEntradaSalida
1AlibabaPRINCIPAL
$0.25
$0.75

Comparar precios entre diferentes proveedores de API para este modelo.

Fuentes externas