Video Question Answering on Video-MME w.o. sub. Short
80AccuracyQwen2-VL + ReQuest
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen2-VL + ReQuestLLM Size=8B, #Frames=≤ 512, zero-shot transfer=true2026.07 | 80 | |
| Qwen2-VLLLM Size=8B, #Frames=512, reproduced=true2026.07 | 78.6 | |
| LLaVA-Video + ReQuestLLM Size=7B, #Frames=322026.07 | 77 | |
| LLaVA-VideoLLM Size=7B, #Frames=32, reproduced=true2026.07 | 76.2 | |
| NVILALLM Size=8B, #Frames=10242026.07 | 75 | |
| LLaVa-OneVision + ReQuestLLM Size=7B, #Frames=32, zero-shot transfer=true2026.07 | 71.7 | |
| LLaVa-OneVisionLLM Size=7B, #Frames=32, reproduced=true2026.07 | 70.3 | |
| VideoTreeLLM Size=–, #Frames=–2026.07 | 67.8 | |
| Frame-VoyagerLLM Size=7B, #Frames=82026.07 | 67.3 | |
| LongVULLM Size=7B, #Frames=1fps2026.07 | 64.7 | |
| LLoViLLM Size=–, #Frames=–2026.07 | 62.1 | |
| VideoLLaMA2LLM Size=7B, #Frames=162026.07 | 56 | |
| VideoChat2LLM Size=7B, #Frames=162026.07 | 48.3 | |
| ShareGPT4VideoLLM Size=8B, #Frames=162026.07 | 48.3 | |
| Chat-UniVi-V1.5LLM Size=7B, #Frames=642026.07 | 45.7 | |
| Video-LLaVALLM Size=7B, #Frames=82026.07 | 45.3 |