Qwen2.5-VL Technical Report

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Bai, Shuai, Chen, Keqin, Liu, Xuejing, Wang, Jialin, Ge, Wenbin, Song, Sibo, Dang, Kai, Wang, Peng, Wang, Shijie, Tang, Jun, Zhong, Humen, Zhu, Yuanzhi, Yang, Mingkun, Li, Zhaohai, Wan, Jianqiang, Wang, Pengfei, Ding, Wei, Fu, Zheren, Xu, Yiheng, Ye, Jiabo, Zhang, Xi, Xie, Tianbao, Cheng, Zesen, Zhang, Hang, Yang, Zhibo, Xu, Haiyang, Lin, Junyang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913718380527616
author Bai, Shuai
Chen, Keqin
Liu, Xuejing
Wang, Jialin
Ge, Wenbin
Song, Sibo
Dang, Kai
Wang, Peng
Wang, Shijie
Tang, Jun
Zhong, Humen
Zhu, Yuanzhi
Yang, Mingkun
Li, Zhaohai
Wan, Jianqiang
Wang, Pengfei
Ding, Wei
Fu, Zheren
Xu, Yiheng
Ye, Jiabo
Zhang, Xi
Xie, Tianbao
Cheng, Zesen
Zhang, Hang
Yang, Zhibo
Xu, Haiyang
Lin, Junyang
author_facet Bai, Shuai
Chen, Keqin
Liu, Xuejing
Wang, Jialin
Ge, Wenbin
Song, Sibo
Dang, Kai
Wang, Peng
Wang, Shijie
Tang, Jun
Zhong, Humen
Zhu, Yuanzhi
Yang, Mingkun
Li, Zhaohai
Wan, Jianqiang
Wang, Pengfei
Ding, Wei
Fu, Zheren
Xu, Yiheng
Ye, Jiabo
Zhang, Xi
Xie, Tianbao
Cheng, Zesen
Zhang, Hang
Yang, Zhibo
Xu, Haiyang
Lin, Junyang
contents We introduce Qwen2.5-VL, the latest flagship model of Qwen vision-language series, which demonstrates significant advancements in both foundational capabilities and innovative functionalities. Qwen2.5-VL achieves a major leap forward in understanding and interacting with the world through enhanced visual recognition, precise object localization, robust document parsing, and long-video comprehension. A standout feature of Qwen2.5-VL is its ability to localize objects using bounding boxes or points accurately. It provides robust structured data extraction from invoices, forms, and tables, as well as detailed analysis of charts, diagrams, and layouts. To handle complex inputs, Qwen2.5-VL introduces dynamic resolution processing and absolute time encoding, enabling it to process images of varying sizes and videos of extended durations (up to hours) with second-level event localization. This allows the model to natively perceive spatial scales and temporal dynamics without relying on traditional normalization techniques. By training a native dynamic-resolution Vision Transformer (ViT) from scratch and incorporating Window Attention, we reduce computational overhead while maintaining native resolution. As a result, Qwen2.5-VL excels not only in static image and document understanding but also as an interactive visual agent capable of reasoning, tool usage, and task execution in real-world scenarios such as operating computers and mobile devices. Qwen2.5-VL is available in three sizes, addressing diverse use cases from edge AI to high-performance computing. The flagship Qwen2.5-VL-72B model matches state-of-the-art models like GPT-4o and Claude 3.5 Sonnet, particularly excelling in document and diagram understanding. Additionally, Qwen2.5-VL maintains robust linguistic performance, preserving the core language competencies of the Qwen2.5 LLM.
format Preprint
id arxiv_https___arxiv_org_abs_2502_13923
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Qwen2.5-VL Technical Report
Bai, Shuai
Chen, Keqin
Liu, Xuejing
Wang, Jialin
Ge, Wenbin
Song, Sibo
Dang, Kai
Wang, Peng
Wang, Shijie
Tang, Jun
Zhong, Humen
Zhu, Yuanzhi
Yang, Mingkun
Li, Zhaohai
Wan, Jianqiang
Wang, Pengfei
Ding, Wei
Fu, Zheren
Xu, Yiheng
Ye, Jiabo
Zhang, Xi
Xie, Tianbao
Cheng, Zesen
Zhang, Hang
Yang, Zhibo
Xu, Haiyang
Lin, Junyang
Computer Vision and Pattern Recognition
Computation and Language
We introduce Qwen2.5-VL, the latest flagship model of Qwen vision-language series, which demonstrates significant advancements in both foundational capabilities and innovative functionalities. Qwen2.5-VL achieves a major leap forward in understanding and interacting with the world through enhanced visual recognition, precise object localization, robust document parsing, and long-video comprehension. A standout feature of Qwen2.5-VL is its ability to localize objects using bounding boxes or points accurately. It provides robust structured data extraction from invoices, forms, and tables, as well as detailed analysis of charts, diagrams, and layouts. To handle complex inputs, Qwen2.5-VL introduces dynamic resolution processing and absolute time encoding, enabling it to process images of varying sizes and videos of extended durations (up to hours) with second-level event localization. This allows the model to natively perceive spatial scales and temporal dynamics without relying on traditional normalization techniques. By training a native dynamic-resolution Vision Transformer (ViT) from scratch and incorporating Window Attention, we reduce computational overhead while maintaining native resolution. As a result, Qwen2.5-VL excels not only in static image and document understanding but also as an interactive visual agent capable of reasoning, tool usage, and task execution in real-world scenarios such as operating computers and mobile devices. Qwen2.5-VL is available in three sizes, addressing diverse use cases from edge AI to high-performance computing. The flagship Qwen2.5-VL-72B model matches state-of-the-art models like GPT-4o and Claude 3.5 Sonnet, particularly excelling in document and diagram understanding. Additionally, Qwen2.5-VL maintains robust linguistic performance, preserving the core language competencies of the Qwen2.5 LLM.
title Qwen2.5-VL Technical Report
topic Computer Vision and Pattern Recognition
Computation and Language
url https://arxiv.org/abs/2502.13923