Long-video Understanding on LSDBench
60.35AccuracyInternVL3.5-4B
Evaluation Results
| Method | Links | |
|---|---|---|
| InternVL3.5-4BCheckpoint=Base2026.05 | 60.35 | |
| Video-ZeroBackbone=Qwen3-VL-8B-Instruct, Iteration=12026.05 | 58.13 | |
| Video-Zero (Peak)Backbone=Qwen3-VL-8B-Instruct2026.05 | 58.13 | |
| Base ModelBackbone=Qwen3-VL-8B-Instruct2026.05 | 57.59 | |
| VisPlayBackbone=Qwen3-VL-8B-Instruct, Iteration=52026.05 | 57.59 | |
| Base ModelBackbone=Qwen3-VL-8B-Instruct2026.05 | 57.59 | |
| Video-ZeroBackbone=InternVL3.5-4B, Checkpoint=v12026.05 | 57.52 | |
| VisPlayBackbone=Qwen3-VL-8B-Instruct, Iteration=22026.05 | 57.44 | |
| VisPlayBackbone=Qwen3-VL-8B-Instruct, Iteration=42026.05 | 57.44 | |
| VisPlayBackbone=Qwen3-VL-8B-Instruct2026.05 | 57.44 | |
| VisPlayBackbone=Qwen3-VL-8B-Instruct, Iteration=12026.05 | 57.36 | |
| Video-ZeroBackbone=InternVL3.5-4B, Checkpoint=v52026.05 | 57.36 | |
| V-ZeroBackbone=Qwen3-VL-8B-Instruct, Iteration=22026.05 | 57.29 | |
| Video-ZeroBackbone=Qwen3-VL-8B-Instruct, Iteration=22026.05 | 57.29 | |
| Video-ZeroBackbone=Qwen3-VL-8B-Instruct, Iteration=32026.05 | 57.29 | |
| VisPlayBackbone=Qwen3-VL-8B-Instruct, Iteration=32026.05 | 57.21 | |
| V-ZeroBackbone=Qwen3-VL-8B-Instruct, Iteration=32026.05 | 57.21 | |
| Video-ZeroBackbone=InternVL3.5-4B, Checkpoint=v22026.05 | 57.21 | |
| V-ZeroBackbone=Qwen3-VL-8B-Instruct, Iteration=12026.05 | 57.06 | |
| Video-ZeroBackbone=InternVL3.5-4B, Checkpoint=v32026.05 | 57.06 | |
| Video-ZeroBackbone=InternVL3.5-4B, Checkpoint=v42026.05 | 56.98 | |
| V-ZeroBackbone=Qwen3-VL-8B-Instruct, Iteration=42026.05 | 56.9 | |
| V-ZeroBackbone=Qwen3-VL-8B-Instruct, Iteration=52026.05 | 56.9 | |
| V-ZeroBackbone=Qwen3-VL-8B-Instruct2026.05 | 56.9 | |
| Video-ZeroBackbone=Qwen3-VL-8B-Instruct, Iteration=42026.05 | 56.75 | |
| Video-Zero (Iter 4)Backbone=Qwen3-VL-8B-Instruct2026.05 | 56.75 | |
| Video-ZeroBackbone=Qwen3-VL-4B-Instruct, Iteration=52026.05 | 56.67 | |
| Video-Zero (Peak)Backbone=Qwen3-VL-4B-Instruct2026.05 | 56.67 | |
| Video-ZeroBackbone=Qwen3-VL-4B-Instruct, Iteration=42026.05 | 56.52 | |
| Video-Zero (Iter 4)Backbone=Qwen3-VL-4B-Instruct2026.05 | 56.52 | |
| Video-ZeroBackbone=Qwen3-VL-8B-Instruct, Iteration=52026.05 | 56.06 | |
| Video-ZeroBackbone=MiMo-SFT-7B, Checkpoint=v32026.05 | 55.83 | |
| Video-ZeroBackbone=MiMo-SFT-7B, Checkpoint=v42026.05 | 55.67 | |
| Video-ZeroBackbone=Qwen3-VL-4B-Instruct, Iteration=32026.05 | 55.29 | |
| Video-ZeroBackbone=Qwen3-VL-4B-Instruct, Iteration=22026.05 | 55.21 | |
| Video-ZeroBackbone=MiMo-SFT-7B, Checkpoint=v52026.05 | 55.14 | |
| Video-ZeroBackbone=Qwen3-VL-4B-Instruct, Iteration=12026.05 | 54.98 | |
| VisPlayBackbone=Qwen3-VL-4B-Instruct, Iteration=22026.05 | 54.91 | |
| VisPlayBackbone=Qwen3-VL-4B-Instruct, Iteration=32026.05 | 54.83 | |
| VisPlayBackbone=Qwen3-VL-4B-Instruct, Iteration=42026.05 | 54.83 | |
| V-ZeroBackbone=Qwen3-VL-4B-Instruct, Iteration=52026.05 | 54.83 | |
| VisPlayBackbone=Qwen3-VL-4B-Instruct2026.05 | 54.83 | |
| VisPlayBackbone=Qwen3-VL-4B-Instruct, Iteration=12026.05 | 54.75 | |
| V-ZeroBackbone=Qwen3-VL-4B-Instruct, Iteration=22026.05 | 54.75 | |
| Base ModelBackbone=Qwen3-VL-4B-Instruct2026.05 | 54.68 | |
| V-ZeroBackbone=Qwen3-VL-4B-Instruct, Iteration=42026.05 | 54.68 | |
| Base ModelBackbone=Qwen3-VL-4B-Instruct2026.05 | 54.68 | |
| V-ZeroBackbone=Qwen3-VL-4B-Instruct2026.05 | 54.68 | |
| V-ZeroBackbone=Qwen3-VL-4B-Instruct, Iteration=12026.05 | 54.6 | |
| V-ZeroBackbone=Qwen3-VL-4B-Instruct, Iteration=32026.05 | 54.6 | |
| VisPlayBackbone=Qwen3-VL-4B-Instruct, Iteration=52026.05 | 54.52 | |
| Video-ZeroBackbone=MiMo-SFT-7B, Checkpoint=v12026.05 | 54.45 | |
| Video-ZeroBackbone=MiMo-SFT-7B, Checkpoint=v22026.05 | 54.29 | |
| MiMo-SFT-7BBackbone=MiMo-SFT-7B, Checkpoint=Base2026.05 | 51.92 |