Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Natural Language Visual Reasoning on NLVR2
Loading...
87.3
Accuracy
Ours (masked) (LLaVA-OV-7B)
5.556
26.778
48
69.222
Sep 14, 2023
Feb 2, 2024
Jun 23, 2024
Nov 12, 2024
Apr 3, 2025
Aug 23, 2025
Jan 12, 2026
Accuracy
Updated 2d ago
Evaluation Results
Method
Method
Links
Accuracy
Ours (masked) (LLaVA-OV-7B)
Model Size=7B, Backbon...
2026.01
87.3
LLaVA-OV-7B
Model Size=7B
2026.01
84.2
Ours (LLaVA-OV-1.5B)
Model Size=1.5B, Backb...
2026.01
73.7
LLaVA-OV-1.5B
Model Size=1.5B
2026.01
70.9
Ours (masked) (LLaVA-OV-1.5B)
Model Size=1.5B, Backb...
2026.01
69
Ours (LLaVA-OV-0.5B)
Model Size=0.5B, Backb...
2026.01
68
MMICL
2023.09
66.6
Ours (masked) (LLaVA-OV-0.5B)
Model Size=0.5B, Backb...
2026.01
65.1
LLaVA-OV-0.5B
Model Size=0.5B
2026.01
61.2
InstructionBlip
2023.09
53.95
OTTER
2023.09
47.2
Qwen2VL-2B
Model Size=2B
2026.01
41.6
Qwen2VL-7B
Model Size=7B
2026.01
41.5
InternVL2-2B
Model Size=2B
2026.01
18.9
InternVL2-8B
Model Size=8B
2026.01
8.7
Feedback
Search any
task
Search any
task