Qwen3-VL-235B-A22B
Qwen Team (Alibaba Cloud)
Parameters
—
Architecture
—
Released
—
License
—
About
No detailed description available for this model.
Benchmark Scores
| Benchmark | Score | Date |
|---|---|---|
|
MMMU
vision_language
|
90.39%
|
01.02.2026 |
|
MMMU-Pro
vision_language
|
64.03%
|
01.02.2026 |
|
MathVision
vision_language
|
55.21%
|
01.02.2026 |
|
MathVista (mini)
vision_language
|
77.01%
|
01.02.2026 |
|
DynaMath
vision_language
|
44.94%
|
01.02.2026 |
|
ZEROBench
vision_language
|
2.17%
|
01.02.2026 |
|
ZEROBench_sub
vision_language
|
23.53%
|
01.02.2026 |
|
VlmsAreBlind
vision_language
|
17.45%
|
01.02.2026 |
|
BabyVision
vision_language
|
12.47%
|
01.02.2026 |
|
RealWorldQA
vision_language
|
83.26%
|
01.02.2026 |
|
MMStar
vision_language
|
51.89%
|
01.02.2026 |
|
MMBench EN-DEV v1.1
vision_language
|
48.33%
|
01.02.2026 |
|
SimpleVQA
vision_language
|
75.53%
|
01.02.2026 |
|
HallusionBench
vision_language
|
67.33%
|
01.02.2026 |
|
OmniDocBench 1.5
document_understanding
|
92.23%
|
01.02.2026 |
|
CharXiv (RQ)
document_understanding
|
0.40%
|
01.02.2026 |
|
MMLongBench-Doc
document_understanding
|
59.60%
|
01.02.2026 |
|
OCRBench
document_understanding
|
70.32%
|
01.02.2026 |
|
ERQA
spatial_intelligence
|
37.14%
|
01.02.2026 |
|
CountBench
spatial_intelligence
|
47.44%
|
01.02.2026 |
|
RefCOCO (avg)
spatial_intelligence
|
68.18%
|
01.02.2026 |
|
ODInW13
spatial_intelligence
|
21.65%
|
01.02.2026 |
|
EmbSpatialBench
spatial_intelligence
|
97.66%
|
01.02.2026 |
|
RefSpatialBench
spatial_intelligence
|
99.85%
|
01.02.2026 |
|
LingoQA
spatial_intelligence
|
78.03%
|
01.02.2026 |
|
Hypersim
spatial_intelligence
|
11.00
|
01.02.2026 |
|
SUNRGBD
spatial_intelligence
|
53.57%
|
01.02.2026 |
|
Nuscene
spatial_intelligence
|
72.73%
|
01.02.2026 |
|
VideoMME (w sub.)
video_understanding
|
49.59%
|
01.02.2026 |
|
VideoMME (w/o sub.)
video_understanding
|
43.02%
|
01.02.2026 |
|
VideoMMMU
video_understanding
|
42.75%
|
01.02.2026 |
|
MLVU
video_understanding
|
75.86%
|
01.02.2026 |
|
MVBench
video_understanding
|
76.92%
|
01.02.2026 |
|
LVBench
video_understanding
|
4.41%
|
01.02.2026 |
|
MMVU
video_understanding
|
36.05%
|
01.02.2026 |
|
ScreenSpot Pro
general_agent
|
64.66%
|
01.02.2026 |
|
OSWorld-Verified
general_agent
|
5.08%
|
01.02.2026 |
|
AndroidWorld
general_agent
|
19.69%
|
01.02.2026 |
|
TIR-Bench
vision_language
|
14.77%
|
01.02.2026 |
|
V-Star
vision_language
|
68.59%
|
01.02.2026 |
|
SLAKE
vision_language
|
54.70
|
01.02.2026 |
|
PMC-VQA
vision_language
|
18.15%
|
01.02.2026 |
|
MedXpertQA-MM
vision_language
|
55.02%
|
01.02.2026 |
|
AI2D_TEST
document_understanding
|
86.38%
|
01.02.2026 |
|
CC-OCR
document_understanding
|
96.40%
|
01.02.2026 |