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

Text-based Visual Question Answering on TextVQA (VQA^T)

70.4Accuracy

CogVLM-17B

Updated 3d ago

Evaluation Results

Method	Links
CogVLM-17B 2024.07		70.4
MGM-HD+CAL 2024.05		68.8
SPHINX-MoE 2024.08		68
MGM-HD 2024.05		67.2
MGM-HD+CAL 2024.05		67.1
LLaVA-NeXT+CAL 2024.05		67.1
LLaVA-NeXT 2024.05		67
VILA-13B 2024.07		66.6
Mini-Gemini-13B 2024.07		65.9
MGM-HD 2024.05		65.5
Mini-Gemini-7B 2024.07		65.2
CoS-7B 2024.07		65.1
LLaVA-NeXT+CAL 2024.05		65
VILA-7B 2024.07		64.4
LLaVA-NeXT 2024.05		64.2
Qwen-VL 2024.07		63.8
Qwen-VL-7B 2024.08		63.8
MGM+CAL 2024.05		63.8
Qwen-VL 2025.12		63.8
MGM+CAL 2024.05		63
MGM 2024.05		62.6
Qwen-VL-7B-Chat 2024.08		61.5
LLaVA-1.5-13B 2024.07		61.3
LLaVA-1.5 2024.08		61.3
MGM 2024.05		61.1
MoExtend 2024.08		58.7
HyperLLaVA 2024.08		58.5
LLaVA-1.5-7B 2024.07		58.2
mPLUG-Owl2 2024.07		58.2
LLaVA-1.5 2024.08		58.2
LLaVA-1.5 2025.12		58.2
MoELoRA 2024.08		57.1
TeamLoRA 2024.08		57.1
MoE-LLaVA-2.7Bx4 2024.08		57
LLaVA-13b 2024.06		57
LoRA 2024.08		56.9
LLaVolta 2024.06		56.7
MGM+CAL 2024.05		51.8
MoE-LLaVA-2.7Bx4 2024.08		51.4
InstructBLIP-13B 2024.07		50.7
InstructBLIP-13B 2024.08		50.7
InstructBLIP 2025.12		50.7
InstructBLIP-7B 2024.07		50.1
InstructBLIP-7B 2024.08		50.1
InstructBLIP 2025.12		50.1
LLaVA-1.5+CAL 2024.05		49.6
LLaVA-1.5 2024.05		49
MobileVLM 3B 2025.12		48.58
MGM 2024.05		48.1
LLaVA-1.5 2024.05		47.6
LLaVA-1.5+CAL 2024.05		47.5
HyDRA(Coarse-grained) 2025.12		47.14
HyDRA(Fine-grained) 2025.12		46.91
MobileVLM 3B 2025.12		46.58
BLIP-2 2024.08		42.5
BLIP-2 2025.12		42.5
HyDRA(Coarse-grained) 2025.12		40.98
HyDRA(Fine-grained) 2025.12		40.75
MobileVLM 1.7B 2025.12		40.36
MobileVLM 1.7B 2025.12		40.28
IDEFICS-80B 2024.08		30.9
IDEFICS-80B 2025.12		30.9
IDEFICS-9B 2024.07		25.9
IDEFICS-9B 2024.08		25.9
IDEFICS-9B 2025.12		25.9