List your company
ResearchFormerly WizWand

Streaming Video Understanding on OVOBench

92.8AccuracyHuman
40.165653.830367.49581.1597May 11, 2025Jul 19, 2025Sep 26, 2025Dec 4, 2025Feb 11, 2026Apr 21, 2026Jun 30, 2026
Updated 4mo ago

Evaluation results

MethodLinks
Human2026.0692.8
Seed2.0 Pro2026.0677
Seed2.0 Lite2026.0676.7
Seed1.82026.0672.6
Seed1.5-VLThinking mode=true, Evaluation frame rate=1 FPS2025.0572.3
Seed1.5-VLThinking mode=false, Evaluation frame rate=1 FPS2025.0572
Seed2.0 Mini2026.0670.4
Gemini-3-Pro2026.0670.1
Gemini-3-Flash2026.0668.7
Gemini 1.5-ProEvaluation frame rate=1 FPS2025.0567.7
ViCoStreamBackbone=Qwen2.5-VL-7B, acn=8, ur=64, s=0.25 (↓48.4%)2026.0650.77
ViCoStreamBackbone=Qwen2.5-VL-7B, acn=8, ur=32, s=0.25 (↓48.4%)2026.0650.35
ViCoStreamBackbone=Qwen2.5-VL-7B, acn=4, ur=32, s=0.25 (↓48.0%)2026.0649.71
ViCoStreamBackbone=Qwen2.5-VL-7B, acn=4, ur=64, s=0.25 (↓48.0%)2026.0649.41
Qwen2.5-VL-7BMode=Traditional Streaming Prefill (Baseline), acn=∞, ur=∞, s=–2026.0649.35
ViCoStreamBackbone=Qwen2.5-VL-7B, acn=4, ur=64, s=0.1 (↓70.6%)2026.0647.56
ViCoStreamBackbone=Qwen2.5-VL-7B, acn=8, ur=32, s=0.1 (↓70.5%)2026.0647.32
ViCoStreamBackbone=Qwen2.5-VL-7B, acn=8, ur=64, s=0.1 (↓70.5%)2026.0647.23
Qwen2.5-VL-3BMode=Traditional Streaming Prefill (Baseline), acn=∞, ur=∞, s=–2026.0647.13
ViCoStreamBackbone=Qwen2.5-VL-7B, acn=4, ur=32, s=0.1 (↓70.6%)2026.0646.77
ViCoStreamBackbone=Qwen2.5-VL-3B, acn=8, ur=64, s=0.25 (↓57.3%)2026.0646.35
ViCoStreamBackbone=Qwen2.5-VL-3B, acn=8, ur=32, s=0.25 (↓57.3%)2026.0646.17
ViCoStreamBackbone=Qwen2.5-VL-3B, acn=4, ur=64, s=0.25 (↓57.6%)2026.0645.3
ViCoStreamBackbone=Qwen2.5-VL-3B, acn=4, ur=32, s=0.25 (↓57.6%)2026.0645.16
ViCoStreamBackbone=Qwen2.5-VL-3B, acn=8, ur=32, s=0.1 (↓72.9%)2026.0643.85
ViCoStreamBackbone=Qwen2.5-VL-3B, acn=8, ur=64, s=0.1 (↓73.0%)2026.0643.28
ViCoStreamBackbone=Qwen2.5-VL-3B, acn=4, ur=32, s=0.1 (↓72.9%)2026.0642.48
ViCoStreamBackbone=Qwen2.5-VL-3B, acn=4, ur=64, s=0.1 (↓72.9%)2026.0642.19