Our new X account is live! Follow @wizwand_team for updates

Multi-modal Question Answering on MMBench

86.4Accuracy

Qwen2.5vl-Instruct

Updated 3d ago

Evaluation Results

Method	Links
Qwen2.5vl-Instruct 2025.12		86.4
Qwen2.5vl-Instruct 2025.12		85.1
MergeMix 2025.10		84.19
Qwen2.5-VL-Ins-7B 2025.10		84.02
Qwen3-VL-4B 2026.02		83.9
DenseMLLM-4B 2026.02		83.9
InternVL2.5-8B-VisualPRM 2025.12		83.5
SFT Vision 2025.10		83.41
Qwen2.5vl-Instruct 2025.12		83.3
VisionThink-7B 2025.10		82.73
Cambrian-1 2025.12		81.4
InternVL-3.5-4B 2026.02		80.3
Curr-ReFT 2025.12		79
R1-Onevision 2025.12		75.6
Avg 2026.02		75.52
GPT-4V 2025.12		75
MaD-Mix 2026.02		74.57
Fused 2026.02		73.28
Uniform 2026.02		71.05
HyperSeg 2024.11		67.9
LLaVA-1.5 2024.11		64.3
Qwen-VL-Chat 2024.11		60.6
Shikra 2024.11		58.8
URSA 2025.12		55.5
Qwen-VL 2024.11		38.2
InstructBLIP 2024.11		36
Fused 2026.02		35.14
MaD-Mix 2026.02		34.45
Uniform 2026.02		34.36
Avg 2026.02		26.8