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

Video Question Answering on NEXT-QA

85.5Overall Accuracy

LinVT-Qwen2-VL

Updated 3d ago

Evaluation Results

Method
LinVT-Qwen2-VL 2024.12	85.5	-	-	-	-	-
LLaVA-Video (Chain-of-Thought) 2025.03	84.61	-	-	-	-	-
VITED (Temporal Evidence Distillation) 2025.03	84.13	-	-	-	-	-
LLaVA-Video (Video Instruction Tuning) 2025.03	83.93	-	-	-	-	-
VITED (Dense Caption Distillation) 2025.03	83.91	-	-	-	-	-
LLaVA-Video 2024.12	83.2	-	-	-	-	-
LLaVA-Video 2025.03	83.2	-	-	-	-	-
PVC InternVL2 2024.12	82	-	-	-	-	-
Oryx MLLM 2024.12	81.9	-	-	-	-	-
LinVT-InternVL2 2024.12	81.9	-	-	-	-	-
CoPE-VideoLM 2026.02	81.8	-	-	-	-	-
LinVT-Molmo 2024.12	81.4	-	-	-	-	-
LLaVA-Scissor 2026.02	81.2	-	-	-	-	-
FastV 2026.02	81.1	-	-	-	-	-
DyCoke 2026.02	81.1	-	-	-	-	-
PLLaVA 2026.02	81	-	-	-	-	-
Magma-8B 2025.02	80.9	-	-	-	-	-
LinVT-BLIP-3 2024.12	80.1	-	-	-	-	-
PAVE-7B (w/ video feature) 2025.03	79.6	-	-	-	-	-
VideoChat2-HD 2024.12	79.5	-	-	-	-	-
LLaVA-OneVision 32 frames + BOLT 2025.03	79.5	-	-	-	-	-
LLaVA-OV 2024.12	79.4	-	-	-	-	-
LLAVA-OV 2025.02	79.4	-	-	-	-	-
LLaVA-OneVision 2025.03	79.4	-	-	-	-	-
LLaVA-OneVision 32 frames 2025.03	79.4	-	-	-	-	-
LLaVA-OV-7B 2025.03	79.4	-	-	-	-	-
mPLUG-Owl3 2024.08	78.6	-	-	-	-	-
VisionZip 2026.02	78.5	-	-	-	-	-
LLaVA-OneVision 16 frames + BOLT 2025.03	78.3	-	-	-	-	-
LLAVA-Interleave 2024.08	78.2	-	-	-	-	-
LLaVA-OneVision 16 frames 2025.03	78.1	-	-	-	-	-
LLaVA-OneVision 8 frames 2025.03	77.4	-	-	-	-	-
LLaVA-OneVision 8 frames + BOLT 2025.03	77.4	-	-	-	-	-
Vamos 2023.11	77.3	81.7	75.3	77.2	-	-
BLIP-3-Video 2024.12	77.1	-	-	-	-	-
LinVT-Aquila 2024.12	76.6	-	-	-	-	-
LLAMA-VQA-33B 2023.11	75.5	78.8	72.6	76.2	-	-
Frame-Voyager 2025.03	73.9	-	-	-	-	-
SeViLA 2023.11	73.8	81.3	69.4	74.2	-	-
SeViLA 2023.12	73.8	-	-	-	-	-
SeViLA 2025.03	73.8	-	-	-	-	-
VITED (Temporal Evidence Distillation) 2025.03	73.42	-	-	-	-	-
PVC InternVL2 2024.12	73.4	-	-	-	-	-
LLAMA-VQA-7B 2023.11	72	75.8	69.2	72.7	-	-
VITED (Dense Caption Distillation) 2025.03	71.94	-	-	-	-	-
Tarsier 2024.12	71.6	-	-	-	-	-
LinVT-Mipha 2024.12	71.1	-	-	-	-	-
TimeChat (Video Instruction Tuning) 2025.03	71.05	-	-	-	-	-
IXC-2.5 2024.12	71	-	-	-	-	-
LLaVA-NeXT-Video 2024.12	70.2	-	-	-	-	-
BLIP-2 2023.11	70.1	80.1	65.2	70.1	-	-
LongVA 2025.02	69.3	-	-	-	-	-
VideoChat2 2024.08	68.6	-	-	-	-	-
LongVA 2024.12	68.3	-	-	-	-	-
LongVA 2025.03	68.3	-	-	-	-	-
LLaMA-3.2V 2025.03	67.58	-	-	-	-	-
ATP 2023.12	65.8	50.2	53.1	-	-	-
InternVideo2 2025.03	63.4	-	-	-	-	-
RTQ 2023.12	63.2	75.6	59.6	61.4	-	-
Intern Video 2023.11	63.2	75.8	58.5	62.5	-	-
HiTeA 2023.11	63.1	75.6	58.3	62.4	-	-
BLIPv2 ViTG FlanT5xxl 2023.09	62.4	-	-	-	-	-
HiTeA 2023.12	62.4	75.5	58.7	60.6	-	-
DeepStack-L-7B 2024.06	61	69.4	55.5	61.9	-	-
COVGT 2023.12	59.7	68.4	58	58	-	-
LLaVA-1.5-7B 2024.06	59.6	68.9	55.5	59.5	-	-
COVGT 2023.12	59.4	66.8	57	58.5	-	-
VFC 2023.04	58.6	72.8	53.3	57.6	-	-
AnyMAL-Image 70B 2023.09	57.6	-	-	-	-	-
OneLLM-7B 2023.12	57.3	-	-	-	-	-
LLaVA-OV-0.5B 2025.03	57.2	-	-	-	-	-
PAVE-0.5B (w/ video feature) 2025.03	56.1	-	-	-	-	-
VGT 2023.04	55	64.1	55.1	52.3	-	-
Kangaroo 2024.12	54.8	-	-	-	-	-
TAATP 2023.04	54.3	66.8	50.2	53.1	-	-
ATP 2023.11	54.3	66.8	50.2	53.1	-	-
JustAsk 2023.12	52.3	-	-	-	-	-
HQGA 2023.12	51.8	59.4	52.3	49	-	-
HGQA 2023.12	51.8	-	-	-	-	-
VFC 2023.04	51.5	64.1	45.4	51.6	-	-
Temp[ATP] 2023.04	51.5	65	49.3	48.6	-	-
Video-Llava 2025.02	51.4	-	-	-	-	-
IGV 2023.12	51.3	59.6	51.7	48.6	-	-
AnyMAL-Video 70B 2023.09	50.6	-	-	-	-	-
TimeChat 2024.12	50.6	-	-	-	-	-
HGA 2023.04	49.7	59.3	50.7	46.3	-	-
ATP 2023.04	49.2	58.9	46.7	48.3	-	-
Internvideo 2023.09	49.1	-	-	-	-	-
InternVideo 2023.12	49.1	-	-	-	-	-
AnyMAL-Image 13B 2023.09	47.9	-	-	-	-	-
AnyMAL-13B 2023.12	47.9	-	-	-	-	-
AnyMAL-Video 13B 2023.09	46.8	-	-	-	-	-
CLIP 2023.04	43.9	57	38.1	43.6	-	-
Video-Chat2 2025.02	43.3	-	-	-	-	-
PaLI-X 2023.05	38.3	-	-	-	-	-
PALI-X-55B 2023.07	38.3	-	-	-	-	-
PaLI-X 2023.05	37	-	-	-	-	-
Flamingo 2023.05	33.5	-	-	-	-	-
TimeChat (Chain-of-Thought) 2025.03	25.6	-	-	-	-	-
Emu 2023.07	23.4	-	-	-	-	-

Showing 100 of 139 rows