List your company

Long-video Understanding on VideoNIAH

46.81AccuracyVideo-Zero
36.025238.825141.62544.4249May 14, 2026

Evaluation results

MethodLinks
Video-ZeroBackbone=MiMo-SFT-7B, Checkpoint=v12026.0546.81
Video-ZeroBackbone=MiMo-SFT-7B, Checkpoint=v22026.0546.74
MiMo-SFT-7BBackbone=MiMo-SFT-7B, Checkpoint=Base2026.0546.67
Video-ZeroBackbone=Qwen3-VL-4B-Instruct, Iteration=42026.0546.37
Video-ZeroBackbone=Qwen3-VL-8B-Instruct, Iteration=42026.0546.37
Video-Zero (Iter 4)Backbone=Qwen3-VL-4B-Instruct2026.0546.37
Video-Zero (Peak)Backbone=Qwen3-VL-4B-Instruct2026.0546.37
Video-Zero (Iter 4)Backbone=Qwen3-VL-8B-Instruct2026.0546.37
Video-Zero (Peak)Backbone=Qwen3-VL-8B-Instruct2026.0546.37
Video-ZeroBackbone=Qwen3-VL-8B-Instruct, Iteration=52026.0546.15
Video-ZeroBackbone=Qwen3-VL-4B-Instruct, Iteration=52026.0546.07
V-ZeroBackbone=Qwen3-VL-8B-Instruct, Iteration=42026.0546.07
Video-ZeroBackbone=Qwen3-VL-8B-Instruct, Iteration=22026.0546.07
V-ZeroBackbone=Qwen3-VL-8B-Instruct2026.0546.07
Video-ZeroBackbone=Qwen3-VL-8B-Instruct, Iteration=32026.0545.93
VisPlayBackbone=Qwen3-VL-8B-Instruct, Iteration=12026.0545.85
VisPlayBackbone=Qwen3-VL-8B-Instruct, Iteration=52026.0545.85
Video-ZeroBackbone=Qwen3-VL-8B-Instruct, Iteration=12026.0545.85
Base ModelBackbone=Qwen3-VL-4B-Instruct2026.0545.78
VisPlayBackbone=Qwen3-VL-4B-Instruct, Iteration=12026.0545.78
Video-ZeroBackbone=Qwen3-VL-4B-Instruct, Iteration=32026.0545.78
V-ZeroBackbone=Qwen3-VL-8B-Instruct, Iteration=32026.0545.78
Base ModelBackbone=Qwen3-VL-4B-Instruct2026.0545.78
Video-ZeroBackbone=Qwen3-VL-4B-Instruct, Iteration=12026.0545.7
Video-ZeroBackbone=Qwen3-VL-4B-Instruct, Iteration=22026.0545.7
Base ModelBackbone=Qwen3-VL-8B-Instruct2026.0545.7
VisPlayBackbone=Qwen3-VL-8B-Instruct, Iteration=32026.0545.7
VisPlayBackbone=Qwen3-VL-8B-Instruct, Iteration=42026.0545.7
V-ZeroBackbone=Qwen3-VL-8B-Instruct, Iteration=22026.0545.7
V-ZeroBackbone=Qwen3-VL-8B-Instruct, Iteration=52026.0545.7
Base ModelBackbone=Qwen3-VL-8B-Instruct2026.0545.7
VisPlayBackbone=Qwen3-VL-8B-Instruct2026.0545.7
V-ZeroBackbone=Qwen3-VL-4B-Instruct, Iteration=12026.0545.63
V-ZeroBackbone=Qwen3-VL-4B-Instruct, Iteration=22026.0545.63
V-ZeroBackbone=Qwen3-VL-4B-Instruct, Iteration=32026.0545.63
VisPlayBackbone=Qwen3-VL-8B-Instruct, Iteration=22026.0545.63
V-ZeroBackbone=Qwen3-VL-8B-Instruct, Iteration=12026.0545.63
VisPlayBackbone=Qwen3-VL-4B-Instruct, Iteration=32026.0545.56
V-ZeroBackbone=Qwen3-VL-4B-Instruct, Iteration=42026.0545.56
V-ZeroBackbone=Qwen3-VL-4B-Instruct2026.0545.56
V-ZeroBackbone=Qwen3-VL-4B-Instruct, Iteration=52026.0545.48
Video-ZeroBackbone=MiMo-SFT-7B, Checkpoint=v42026.0545.41
VisPlayBackbone=Qwen3-VL-4B-Instruct, Iteration=22026.0545.33
VisPlayBackbone=Qwen3-VL-4B-Instruct, Iteration=52026.0545.33
VisPlayBackbone=Qwen3-VL-4B-Instruct, Iteration=42026.0545.26
Video-ZeroBackbone=MiMo-SFT-7B, Checkpoint=v32026.0545.26
VisPlayBackbone=Qwen3-VL-4B-Instruct2026.0545.26
Video-ZeroBackbone=MiMo-SFT-7B, Checkpoint=v52026.0544.96
InternVL3.5-4BCheckpoint=Base2026.0538.81
Video-ZeroBackbone=InternVL3.5-4B, Checkpoint=v12026.0536.89
Video-ZeroBackbone=InternVL3.5-4B, Checkpoint=v22026.0536.89
Video-ZeroBackbone=InternVL3.5-4B, Checkpoint=v32026.0536.81
Video-ZeroBackbone=InternVL3.5-4B, Checkpoint=v42026.0536.67
Video-ZeroBackbone=InternVL3.5-4B, Checkpoint=v52026.0536.44