List your company
ResearchFormerly WizWand

Visual Perception on V*

89ScoreStep-GUI-8B
4.884826.722448.5670.3976Dec 14, 2025Jan 10, 2026Feb 7, 2026Mar 6, 2026Apr 3, 2026Apr 30, 2026May 28, 2026
Updated 5mo ago

Evaluation results

MethodLinks
Step-GUI-8BParameters=8B2025.1289
VL-Scaler-MiMOSize=7B2026.0589
DeepEyesSize=7B2026.0588.9
VL-ScalerSize=7B2026.0586.6
Qwen3-VL 8B-InstructParameters=8B, Type=Instruct2025.1286.4
Qwen3-VL-30B A3B-InstructParameters=30B A3B, Type=Instruct2025.1286.4
Qwen2.5-VL-72BParameter Scale=72B2025.1285.9
Pixel ReasonerSize=7B2026.0584.3
Qwen3-VL-8B-Thinking + DiGParameter Scale=8B, Thinking Mode=true, Grounding Strategy=DiG2025.1281.2
Qwen2.5-VL-InstructSize=72B2026.0581.2
Qwen2.5-VL-7BModel Category=Specialist VLMs2026.0480.1
Step-GUI-4BParameters=4B2025.1279.1
Qwen3-VL-4B-ThinkingParameter Scale=4B, Thinking Mode=true2025.1279.1
Qwen3-VL-4B-Thinking + DiGParameter Scale=4B, Thinking Mode=true, Grounding Strategy=DiG2025.1279.1
Qwen3-VL-8B-ThinkingParameter Scale=8B, Thinking Mode=true2025.1279.1
MiMO-VL-InstructSize=7B2026.0578.6
Qwen2.5-VL-7BParameter Scale=7B2025.1277
InternVL2.5-78BParameter Scale=78B2025.1275.9
Llava-OVSize=7B2026.0572.8
LoMoModel=Qwen3.5-9B, Evaluation Protocol=Standard Evaluation2026.0571.86
Qwen2.5-VL-InstructSize=7B2026.0571.4
VL-RethinkerSize=7B2026.0568.2
Standard SFTModel=Qwen3.5-9B, Evaluation Protocol=Standard Evaluation2026.0568.13
BAGELModel Category=Unified Models2026.0467.5
GPT5-nano minimalVariant=nano minimal2025.1266.5
OpenVLThinkerSize=7B2026.0565.2
LoMoModel=Qwen3.5-9B, Evaluation Protocol=Rendered Evaluation2026.0564.73
mPLUG-Owl3Size=7B2026.0564.5
LLaDA2.0-UniModel Category=Unified Models2026.0461.8
Standard SFTModel=Qwen3.5-9B, Evaluation Protocol=Rendered Evaluation2026.0561.58
R1-VLSize=7B2026.0560.3
LLaDA-oModel Category=Unified Models2026.0457.6
Lumina-DiMOOModel Category=Unified Models2026.0452.7
LoMoModel=LLaVA-OV1.5-8B, Evaluation Protocol=Standard Evaluation2026.0551.7
InternVL-UModel Category=Unified Models2026.0451.3
GPT-4o-miniSize=-2026.0550.8
Standard SFTModel=LLaVA-OV1.5-8B, Evaluation Protocol=Standard Evaluation2026.0547.71
GPT-4oSize=-2026.0545
LLaDA-VModel Category=Specialist VLMs2026.0441.8
DocopilotSize=8B2026.0540.1
LoMoModel=LLaVA-OV1.5-8B, Evaluation Protocol=Rendered Evaluation2026.0536.13
Standard SFTModel=LLaVA-OV1.5-8B, Evaluation Protocol=Rendered Evaluation2026.058.12