Qwen2.5-VL-72B-Instruct

Name: Qwen2.5-VL-72B-Instruct
Author: 阿里巴巴

多模态大模型

Release date: 2025-01-28更新于: 2025-02-20 19:34:281,202

Live demo

Qwen2.5-VL-72B-Instruct is an AI model published by 阿里巴巴, released on 2025-01-28, for 多模态大模型, with 720.0B parameters, and 128K tokens context length, requiring about 144GB storage, under the Tongyi Qianwen RESEARCH LICENSE AGREEMENT license.

Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology

Qwen2.5-VL-72B-Instruct

Model basics

Reasoning traces

Not supported

Thinking modes

Thinking modes not supported

Context length

128K tokens

Max output length

No data

Model type

多模态大模型

Release date

2025-01-28

Model file size

144GB

MoE architecture

Total params / Active params

720.0B / N/A

Knowledge cutoff

No data

Qwen2.5-VL-72B-Instruct

Benchmark Results

Qwen2.5-VL-72B-Instruct currently shows benchmark results led by DocVQA (1 / 5, score 96.40), MMMU (21 / 28, score 70.20). This page also consolidates core specs, context limits, and API pricing so you can evaluate the model from benchmark results and deployment constraints together.