General Video Understanding on TempCompass
97.3AccuracyHuman
Evaluation results
| Method | Links | |
|---|---|---|
| Human2026.06 | 97.3 | |
| Seed2.0 Pro2026.06 | 89.6 | |
| Gemini-3-Flash2026.06 | 88.3 | |
| Gemini-3-Pro2026.06 | 88 | |
| Seed2.0 Lite2026.06 | 87 | |
| Seed1.82026.06 | 86.9 | |
| Seed2.0 Mini2026.06 | 83.7 | |
| VideoLatent-7B (Ours)Frames=64, Architecture Category=Latent MLLM2026.06 | 74.2 | |
| VideoLatent-7B (Ours)Frames=32, Architecture Category=Latent MLLM2026.06 | 74.1 | |
| VideoLatent-7B (Ours)Frames=16, Architecture Category=Latent MLLM2026.06 | 73.7 | |
| VideoRFT-7BFrames=32, Architecture Category=Standard MLLM2026.06 | 73.7 | |
| Mull-Token-7BFrames=32, Architecture Category=Latent MLLM2026.06 | 73.6 | |
| Mull-Token-7BFrames=64, Architecture Category=Latent MLLM2026.06 | 73.6 | |
| VideoChat-R1Frames=162026.06 | 73.5 | |
| Video-R1-7BFrames=32, Architecture Category=Standard MLLM2026.06 | 73.2 | |
| Video-R1-7BFrames=64, Architecture Category=Standard MLLM2026.06 | 73.2 | |
| Mull-Token-7BFrames=16, Architecture Category=Latent MLLM2026.06 | 73 | |
| VideoCFRFrames=642026.06 | 72.9 | |
| Qwen2.5-VL-7B (CoT)Frames=64, Architecture Category=Standard MLLM, CoT=true2026.06 | 72.9 | |
| VideoCFRFrames=322026.06 | 72.8 | |
| Video-R1-7BFrames=16, Architecture Category=Standard MLLM2026.06 | 72.6 | |
| Qwen2.5-VL-7B (CoT)Frames=32, Architecture Category=Standard MLLM, CoT=true2026.06 | 72.6 | |
| Qwen2.5-VL-7BFrames=32, Architecture Category=Standard MLLM, CoT=false2026.06 | 72.3 | |
| Qwen2.5-VL-7BFrames=64, Architecture Category=Standard MLLM, CoT=false2026.06 | 72.3 | |
| Qwen2.5-VL-7B (CoT)Frames=16, Architecture Category=Standard MLLM, CoT=true2026.06 | 72.2 | |
| LVR-7BFrames=16, Architecture Category=Latent MLLM2026.06 | 72 | |
| LVR-7BFrames=32, Architecture Category=Latent MLLM2026.06 | 72 | |
| LVR-7BFrames=64, Architecture Category=Latent MLLM2026.06 | 72 | |
| Qwen2.5-VL-7BFrames=16, Architecture Category=Standard MLLM, CoT=false2026.06 | 71.6 | |
| Video-R1-7BFrames=162026.06 | 70.8 | |
| VideoCFRFrames=162026.06 | 70.8 | |
| Qwen2.5-VL-7B2026.06 | 69.2 | |
| DeepVideo-R12026.06 | 63.1 | |
| Kangeroo-8B2026.06 | 62.5 | |
| Video-UTR-7B2026.06 | 59.7 | |
| VILA-1.5-8B2026.06 | 58.8 | |
| LongVA-7B2026.06 | 56.9 | |
| MARC-3BFrames=12026.06 | 55.3 | |
| TinyLLaVA-Video-R1Frames=162026.06 | 49.5 | |
| LLaMA-VID2026.06 | 45.6 | |
| Monet-7BFrames=64, Architecture Category=Latent MLLM2026.06 | 34.5 | |
| Monet-7BFrames=32, Architecture Category=Latent MLLM2026.06 | 34.4 | |
| Monet-7BFrames=16, Architecture Category=Latent MLLM2026.06 | 31.2 |