Video understanding models
Models that classify, interpret, or extract information from video
Where it stands
Open models57 products1.00× its peers
Fastest products
1microsoft/xclip-base-patch16-zero-shot · Open models9.3×
2facebook/sam2-hiera-base-plus · Open models9.3×3MCG-NJU/videomae-large-finetuned-kinetics · Open models7.5×4apiantonio/vjepa2.1-vit-large-384 · Open models6.7×
5microsoft/xclip-large-patch14 · Open models6.1×
6SoraExplora/VideoMae · Open models3.8×
7MCG-NJU/videomae-large · Open models3.4×8nvidia/NVILA-8B-HD-Video · Open models2.9×9nvidia/Cosmos-Tokenize1-CV8x8x8-720p · Open models2.9×10facebook/vjepa2-vitl-fpc16-256-ssv2 · Open models2.7×11general-video · Agent skills2.6×
12MCG-NJU/videomae-base-finetuned-ssv2 · Open models2.3×13qubvel-hf/vjepa2-vitl-fpc16-256-ssv2 · Open models2.3×
14DAMO-NLP-SG/VideoLLaMA3-7B · Open models2.0×
15microsoft/xclip-base-patch16 · Open models1.9×
16facebook/cotracker3 · Open models1.8×17TIGER-Lab/VideoScore2 · Open models1.7×
18ggml-org/SmolVLM2-256M-Video-Instruct-GGUF · Open models1.7×
19zai-org/VisionReward-Video · Open models1.6×
20optimum-intel-internal-testing/tiny-random-llava-next-video · Open models1.5×
21DAMO-NLP-SG/VideoLLaMA3-2B · Open models1.5×
22llava-hf/LLaVA-NeXT-Video-7B-hf · Open models1.5×
23google/vivit-b-16x2 · Open models1.4×
24OpenGVLab/VideoMAEv2-Huge · Open models1.4×
25OpenGVLab/InternVideo2_5_Chat_8B · Open models1.4×
83 products carry this theme. “× its peers” is the theme's median growth against the place's median.