List your company
ResearchFormerly WizWand

Perception and Reasoning on RealWorldQA

74.2ScoreQwen3.5-4B
47.607254.511161.41568.3189May 8, 2026May 14, 2026May 20, 2026May 27, 2026Jun 2, 2026Jun 8, 2026Jun 15, 2026
Updated 4mo ago

Evaluation results

MethodLinks
Qwen3.5-4BModel=Qwen3.5-4B2026.0574.2
Molmo2-4BModel=Molmo2-4B2026.0573.8
PLM-3BModel=PLM-3B2026.0573.2
InternVL3.5-20B-A4BModel=InternVL3.5-20B-A4B2026.0571.2
Qwen3-VL-4BModel=Qwen3-VL-4B2026.0571
Qwen3-VL + LOCUSSize=4B2026.0671
DeepSeek-VL2-16B-A2.4BModel=DeepSeek-VL2-16B-A2.4B2026.0570
Qwen3-VLSize=4B2026.0669.5
PixelReasonerVisual Bandwidth Constraint (B)=No Constrain, Tool-calling=True2026.0569.02
Qwen3.5-2BModel=Qwen3.5-2B2026.0569
DeepEyesVisual Bandwidth Constraint (B)=No Constrain, Tool-calling=True2026.0568.89
Qwen2.5-VLVisual Bandwidth Constraint (B)=No Constrain, Reported From=Other publications2026.0568.5
InternVL3.5-4BModel=InternVL3.5-4B2026.0567.8
Starve to PerceiveVisual Bandwidth Constraint (B)=No Constrain, Tool-calling=True2026.0567.71
Qwen2.5-VL + BA-SFTVisual Bandwidth Constraint (B)=No Constrain, Tool-calling=True2026.0567.32
MiMo-VL + LOCUSSize=7B2026.0667.3
GPT-4oVisual Bandwidth Constraint (B)=No Constrain, Reported From=Other publications2026.0567.2
Qwen2.5-VL + BA-SFT + VanillaRLVisual Bandwidth Constraint (B)=No Constrain, Tool-calling=True2026.0567.19
MiMo-VLSize=7B2026.0666.7
Qwen2.5-VL + LOCUSSize=7B2026.0666.4
LLaVA-OneVisionVisual Bandwidth Constraint (B)=No Constrain, Reported From=Other publications2026.0566.3
Qwen3-VL-2BModel=Qwen3-VL-2B2026.0566
Qwen2.5-VL-3BModel=Qwen2.5-VL-3B2026.0565.6
ZAYA1-VL-8B-A1BModel=ZAYA1-VL-8B-A1B2026.0565
Starve to PerceiveVisual Bandwidth Constraint (B)=256, Tool-calling=True2026.0564.97
ChainOfFocusVisual Bandwidth Constraint (B)=No Constrain, Tool-calling=True2026.0564.45
Qwen2.5-VL + BA-SFT + VanillaRLVisual Bandwidth Constraint (B)=256, Tool-calling=True2026.0563.27
DeepEyesVisual Bandwidth Constraint (B)=256, Tool-calling=True2026.0563.01
InternVL3.5-2BModel=InternVL3.5-2B2026.0561.6
Qwen2.5-VLSize=7B2026.0661.6
Qwen2.5-VL + BA-SFTVisual Bandwidth Constraint (B)=256, Tool-calling=True2026.0560.65
MolmoE-8B-A1BModel=MolmoE-8B-A1B2026.0560.4
ChainOfFocusVisual Bandwidth Constraint (B)=256, Tool-calling=True2026.0560
Qwen2.5-VLVisual Bandwidth Constraint (B)=2562026.0556.34
PixelReasonerVisual Bandwidth Constraint (B)=256, Tool-calling=True2026.0555.03
Mini-o3Visual Bandwidth Constraint (B)=No Constrain, Tool-calling=True2026.0552.55
Mini-o3Visual Bandwidth Constraint (B)=256, Tool-calling=True2026.0548.63