List your company
ResearchFormerly WizWand

Fine-grained audio-visual video understanding on OmniVideoBench

58.9AccuracyGemini-2.5-Pro
28.01236.03144.0552.069May 21, 2026May 25, 2026May 30, 2026Jun 4, 2026Jun 8, 2026Jun 13, 2026Jun 18, 2026
Updated 4mo ago

Evaluation results

MethodLinks
Gemini-2.5-ProModel Accessibility=Proprietary2026.0558.9
Gemini-2.0-FlashModel Accessibility=Proprietary2026.0541.5
CogniRoute2026.0638.9
Qwen3 Omni 30BParameters=30B2026.0638.4
LatentOmniModel Accessibility=Open-source, Training/Reasoning Strategy=Latent reasoning2026.0535.4
Qwen2.5-Omni-7B + Explicit Text CoTModel Accessibility=Open-source, Training/Reasoning Strategy=Explicit Text CoT2026.0533.2
OmniVinciModel Accessibility=Open-source2026.0532.1
Baichuan-Omni-1.5Model Accessibility=Open-source2026.0530.7
VITA-1.5-7BModel Accessibility=Open-source2026.0530.5
HumanOmniV2-7BModel Accessibility=Open-source2026.0530.5
Qwen2.5-Omni-7B + Vanilla SFTModel Accessibility=Open-source, Training/Reasoning Strategy=Vanilla SFT2026.0530.5
MiniCPM-o-7BModel Accessibility=Open-source2026.0529.7
Qwen2.5-Omni-7BModel Accessibility=Open-source, Training/Reasoning Strategy=Base2026.0529.3
VideoLLaMA2-7BModel Accessibility=Open-source2026.0529.2