Qwen2.5 VL 32B Instruct
Alibaba Cloud / Qwen TeamQwen開源權重Apache 2.0 · 商用許可
描述
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.
發布日期
2025-02-28
參數規模
33.5B
上下文長度
—
支援模態
—
能力雷達圖
60
general
90
coding
70
reasoning
43
science估算
40
agents
70
multimodal
缺少專門科學評測時，Science 由 LLM Stats 科學得分或推理能力估算。
排行榜排名
基準測試分數 (LLM Stats)
(LLM Stats (zeroeval))Agents
AITZ_EM
83.1%自報
AndroidWorld_SR
22.0%自報
OSWorld
5.9%自報
Biology
GPQANYU + Cohere + Anthropic (2023)
46.0%自報
Code
HumanEvalOpenAI (2021)
91.5%自報
Finance
MMLU
78.4%自報
MMLU-Pro
68.8%自報
General
MBPP
0.84 / 100自報
MMMU
70.0%自報
MMStar
69.5%自報
MMMU-Pro
49.5%自報
Grounding
ScreenSpot
88.5%自報
ScreenSpot Pro
39.4%自報
Image To Text
DocVQADocVQA (2020)
94.8%自報
OCRBench-V2 (zh)
59.1%自報
OCRBench-V2 (en)
57.2%自報
Language
CharadesSTA
54.2%自報
Long Context
LVBench
49.0%自報
Math
MATH
82.2%自報
MathVista-Mini
74.7%自報
MathVision
38.4%自報
Multimodal
Android Control Low_EM
93.3%自報
InfoVQA
83.4%自報
VideoMME w sub.
77.9%自報
CC-OCR
77.1%自報
VideoMME w/o sub.
70.5%自報
Android Control High_EM
69.6%自報
MMBench-Video
1.9%自報
AA 評測指數
(Artificial Analysis)暫無 AA 評測資料
LLM Stats 分類評分
(LLM Stats (zeroeval))Code90
Structured Output80
Text-to-image80
Legal70
Math70
Image To Text70
Language70
Finance70
Healthcare70
Multimodal60
Reasoning60
Spatial Reasoning60
General60
Grounding60
Vision60
Long Context50
Physics50
Biology50
Chemistry50
Video50
Agents40
定價
暫無定價資料
速度
暫無速度資料
供應商價格排行
暫無提供商資料