List your company

Multimodal Reasoning on RealWorldQA

70.46Mean@8 AccuracyQwen2.5-VL-VIF
30.284840.714951.14561.5751May 10, 2026May 11, 2026May 12, 2026May 14, 2026May 15, 2026May 16, 2026May 18, 2026

Evaluation results

MethodLinks
Qwen2.5-VL-VIFParams=7B, Base model=Qwen-2.5-VL2026.0570.46
CAVE-7BParameters=7B2026.0569.7
RAPOBase Model=Qwen3-VL-8B-Instruct2026.0569.54
PAPOBase Model=Qwen3-VL-8B-Instruct2026.0569.41
Qwen3-VL-8B-InstructParameters=8B, Instruct=true2026.0569.3
Qwen2.5-VLParams=7B, Base model=Qwen-2.5-VL2026.0568.63
DeepEyes-7BParameters=7B2026.0568.4
Qwen2.5-VL-SFTParams=7B, Base model=Qwen-2.5-VL2026.0568.37
GRPOBase Model=Qwen3-VL-8B-Instruct2026.0567.66
VPPOBase Model=Qwen3-VL-8B-Instruct2026.0567.48
BaseBase Model=Qwen3-VL-8B-Instruct2026.0567.2
RAPOBase Model=Qwen3-VL-2B-Instruct2026.0565.36
RAPO_DBase Model=Qwen2-VL-7B-Instruct2026.0565.01
RAPO_GBase Model=Qwen2-VL-7B-Instruct2026.0564.67
GRPOBase Model=Qwen3-VL-2B-Instruct2026.0564.28
Cambrian-1Params=8B2026.0564.2
PAPOBase Model=Qwen3-VL-2B-Instruct2026.0564.05
VPPOBase Model=Qwen3-VL-2B-Instruct2026.0563.14
GPT-4V-1106Params=-2026.0563
MM-EurekaBase Model=Qwen2.5-VL-7B-Instruct2026.0561.96
InternVL3.5-8BParameters=8B2026.0561.6
BaseBase Model=Qwen3-VL-2B-Instruct2026.0560.47
Gemini ProParams=-2026.0560.4
mPLUG-Owl3Params=8B2026.0560.4
RAPO_GBase Model=Qwen2.5-VL-7B-Instruct2026.0560.26
Qwen2.5-VLParameters=7B2026.0559.4
VL-RethinkerBase Model=Qwen2.5-VL-7B-Instruct2026.0559.22
RAPO_DBase Model=Qwen2.5-VL-7B-Instruct2026.0558.4
LLaVA-1.5-VIFParams=7B, Base model=LLaVA-1.52026.0557.78
LLaVA-1.5-VIFParams=13B, Base model=LLaVA-1.52026.0556.99
R1-ShareVLBase Model=Qwen2.5-VL-7B-Instruct2026.0556.47
VISTAParams=7B, Base model=LLaVA-1.52026.0556.34
LLaVA-1.5Params=7B, Base model=LLaVA-1.52026.0555.95
LLaVA-1.5-SFTParams=7B, Base model=LLaVA-1.52026.0555.56
LLaVA-1.5-SFTParams=13B, Base model=LLaVA-1.52026.0555.29
LLaVA-1.5Params=13B, Base model=LLaVA-1.52026.0554.9
BaseBase Model=Qwen2.5-VL-7B-Instruct2026.0554.51
TVCBase Model=Qwen2-VL-7B-Instruct2026.0552.88
MINT-CoTBase Model=Qwen2-VL-7B-Instruct2026.0549.15
BaseBase Model=Qwen2-VL-7B-Instruct2026.0531.83