Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Dense Parsing on ScreenParse (test)
Loading...
60.6
Page IoU
ScreenVLM
9.12
22.485
35.85
49.215
Feb 15, 2026
Page IoU
Label Page IoU
mAP@50
Updated 4d ago
Evaluation Results
Method
Method
Links
Page IoU
Label Page IoU
mAP@50
ScreenVLM
Size=316M, Fine-tuned=...
2026.02
60.6
19.7
30.3
RT-DETRv2
Size=43M, Fine-tuned=t...
2026.02
60
17.2
36.2
Qwen3-VL-2B-Instruct
Size=2B, Fine-tuned=tr...
2026.02
58.5
16.6
15.2
YOLO
Size=25M, Fine-tuned=t...
2026.02
53.3
13.3
29.9
InternVL3-2B
Size=2B, Fine-tuned=tr...
2026.02
50.9
17.4
7.2
OmniParser V2
Size=20M, Fine-tuned=t...
2026.02
50.3
14.1
25.1
Qwen3-VL-8B-Instruct
Size=8B, Fine-tuned=false
2026.02
29.4
-
-
OmniParser V2
Size=20M, Fine-tuned=f...
2026.02
27
-
-
Qwen3-VL-2B-Instruct
Size=2B, Fine-tuned=false
2026.02
22.8
5.1
2.3
InternVL3-2B
Size=2B, Fine-tuned=false
2026.02
11.1
3
0
Feedback
Search any
task
Search any
task