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

Visual Question Answering on OK-VQA (Accuracy)

84.7Accuracy

VPD (55B)

Updated 2d ago

Evaluation Results

Method	Links
VPD (55B) 2024.02		84.7
Prophet 2024.02		76.6
Promptcap 2024.02		73.2
Single Task SoTA 2023.12		66.8
Uni-MoE w/ MoE-Task4 (8E) + Aux loss 2024.05		66.4
Uni-MoE w/ MoE-Task4 (8E) 2024.05		66.2
Uni-MoE w/ MoE-Task4 (4E) + Aux loss 2024.05		66.13
PALI-X-55B 2023.08		66.1
Uni-MoE w/ MoE-Task4 (4E) 2024.05		66.02
Emu2-Chat 2023.12		64.8
CogVLM 2024.08		64.8
PaLI-X-VPD 2023.12		64.6
PaLI-17B 2023.11		64.5
PaLI-X-Instruct 2023.12		64.3
Single-Modality-Expert-Task2 2024.05		62.91
AdaMMS 2025.03		62.11
MoE-LLaVA-2.7B×4-Top2 2024.05		62.1
Uni-MoE w/ MoE-Task2 (4E) 2024.05		62.1
LLaVA-1.5 2023.11		62
Qwen-VL-7B 2024.06		62
Qwen-VL-7B + MTV 2024.06		62
Uni-MoE w/ MoE-Task3 (4E) 2024.05		61.96
LLaVA-1.5-13B 2024.05		61.93
Qwen-VL-7B 2024.06		61.5
InfMLLM-7B 2023.11		61.33
DARE-Linear 2025.03		61.07
InstructBlip 7B 2023.10		61
mPLUG-Owl2(base) 2025.03		60.98
Task Arithmetic 2025.03		60.93
CogVLM 2025.03		60.82
mPLUG-Owl3 2024.08		60.1
MoE-LLaVA-1.6B×4-Top2 2024.05		59.9
CVLM 2024.02		58.91
CogVLM 2023.12		58.9
Qwen-VL 2023.08		58.6
QWen-VL-7B 2023.11		58.6
Qwen-VL 2023.12		58.6
Qwen-VL 2023.12		58.6
Qwen-VL-7B 2024.06		58.6
Qwen-VL-7B‡ 2024.05		58.6
Flamingo 2023.12		57.8
Single-Modality-Expert-Task6 2024.05		57.77
mPLUG-Owl2 2023.11		57.7
mPLUG-Owl2 2023.12		57.7
mPLUG-Owl2 2023.12		57.7
mPLUG-Owl2 2024.08		57.7
Uni-MoE w/ MoE-Task1 (4E) 2024.05		57.63
PaLI-3-VPD 2023.12		57.5
InstructBLIP 2024.02		57.36
CVLM (3M IKPairs, Objects=3) 2024.02		57.17
Qwen-VL 2024.02		57.13
CVLM (3M IKPairs, Objects=5) 2024.02		56.92
MiniGPT-v2-7B 2023.11		56.9
CVLM (3M IKPairs, Objects=1) 2024.02		56.9
CVLM (3M IKPairs, Objects=8) 2024.02		56.71
PaLI-3-Instruct 2023.12		56.7
Qwen-VL-Chat 2023.08		56.6
Qwen-VL-Chat 2023.11		56.6
Qwen-VL-Chat 2023.12		56.6
Qwen-VL-13B-Chat 2023.12		56.6
Qwen-VL-Chat 2024.08		56.6
Single-Modality-Expert-Task5 2024.05		56.01
CVLM (3M IKPairs) w/o FKA 2024.02		55.7
PaLM-E-12B 2023.11		55.5
UIO-2XXL 2023.12		55.5
H-GIVR 2026.02		55.43
Mantis-SigLIP 2024.08		55.4
LLaVA-1.5-7B 2024.05		55.09
LLaVA 2023.10		54.4
CVLM 2024.02		54.3
Ties-Merging 2025.03		54.15
Self-Con 2026.02		54.1
Unified-IO-XL 2023.08		54
Unified-IOXL 2023.11		54
UIO-2XL 2023.12		53.7
BLIP-2(FlanT5-xxl)-11B 2024.05		53.7
ViTCoP 2026.01		53.6
Idefics2 2024.08		53.5
Vanilla (LLaVA-1.5-7B) 2026.01		53.4
ViTCoP 2026.01		53.1
Idefics2 + MTV 2024.06		53
CVLM w/o FKA 2024.02		52.8
MANTIS-LLaMA3-8B + MTV 2024.06		52.8
MANTIS-LLaMA3-8B 2024.06		52.5
VisionZip 2026.01		52.5
VK-OOD 2023.02		52.4
Idefics2 2024.06		52.4
Idefics2 2024.06		52.3
MANTIS-LLaMA3-8B 2024.06		52
SparseVLM 2026.01		52
MANTIS-LLaMA3-8B 2024.06		51.7
DARE-Ties 2025.03		51.56
Idefics2 2024.06		51.5
FastV 2026.01		51.2
VK-OOD 2023.02		51.1
Flamingo-9B 2023.08		51
SparseVLM 2026.01		50.9
LLaVA-v1.5† 2024.02		50.8
PyramidDrop 2026.01		50.8
ViTCoP 2026.01		50.8

Showing 100 of 224 rows