Qwen2.5 VL 7B Instruct
Alibaba Cloud / Qwen TeamQwenOpen WeightApache 2.0 · Commercial OK
Description
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.
Release Date
2025-01-26
Parameters
8.3B
Context Length
—
Modalities
—
Capability Radar
60
general
0
coding
50
reasoning
51
scienceest.
50
agents
90
multimodal
Science is estimated from LLM Stats science scores or reasoning when no dedicated science benchmarks are available.
Rankings
| Domain | #Rank | Score | Source |
|---|---|---|---|
| Multimodal Ranking | 138 | 18.0 | LS |
Benchmark Scores (LLM Stats)
(LLM Stats (zeroeval))General
AndroidWorld_SR
25.5%SR
Math
MathVista-Mini
68.2%SR
MathVision
25.1%SR
Multimodal
MobileMiniWob++_SR
91.4%SR
Android Control Low_EM
91.4%SR
AITZ_EM
81.9%SR
VideoMME w/o sub.
65.1%SR
Android Control High_EM
60.1%SR
MMMU
58.6%SR
Reasoning
ChartQAMasry et al. (2022)
87.3%SR
Video
PerceptionTest
70.5%SR
MLVU
70.2%SR
CharadesSTA
43.6%SR
MMBench-Video
1.8%SR
Vision
DocVQADocVQA (2020)
95.7%SR
OCRBench
86.4%SR
TextVQA
84.9%SR
ScreenSpot
84.7%SR
MMBench
84.3%SR
InfoVQA
82.6%SR
CC-OCR
77.8%SR
TempCompass
71.7%SR
VideoMME w sub.
71.6%SR
MVBench
69.6%SR
MMVet
67.1%SR
MMStar
63.9%SR
MMT-Bench
63.6%SR
LongVideoBench
54.7%SR
Hallusion Bench
52.9%SR
LVBench
45.3%SR
MMMU-Pro
38.3%SR
ScreenSpot Pro
29.0%SR
AA Evaluation Indices
(Artificial Analysis)No AA evaluation data available
LLM Stats Category Scores
(LLM Stats (zeroeval))Image To Text90
Structured Output80
Text-to-image80
Long Context60
Multimodal60
Reasoning60
Spatial Reasoning60
General60
Grounding60
Healthcare60
Vision60
Math50
Agents50
Video50
Language40
Pricing
No pricing data available
Speed
No speed data available
Provider Price Ranking
No provider data available