Video Question Answering on Video-MME w.o. sub. Overall
71.1AccuracyQwen2-VL + ReQuest
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen2-VL + ReQuestLLM Size=8B, #Frames=≤ 512, zero-shot transfer=true2026.07 | 71.1 | |
| Qwen2-VLLLM Size=8B, #Frames=512, reproduced=true2026.07 | 70 | |
| LLaVA-Video + ReQuestLLM Size=7B, #Frames=322026.07 | 65.6 | |
| NVILALLM Size=8B, #Frames=10242026.07 | 64 | |
| LLaVA-VideoLLM Size=7B, #Frames=32, reproduced=true2026.07 | 62.6 | |
| LongVULLM Size=7B, #Frames=1fps2026.07 | 60.9 | |
| LLaVa-OneVision + ReQuestLLM Size=7B, #Frames=32, zero-shot transfer=true2026.07 | 60.9 | |
| VideoTreeLLM Size=–, #Frames=–2026.07 | 60.6 | |
| LLaVa-OneVisionLLM Size=7B, #Frames=32, reproduced=true2026.07 | 58.7 | |
| Frame-VoyagerLLM Size=7B, #Frames=82026.07 | 57.5 | |
| LLoViLLM Size=–, #Frames=–2026.07 | 54.7 | |
| VideoLLaMA2LLM Size=7B, #Frames=162026.07 | 47.9 | |
| TimeSuiteLLM Size=7B, #Frames=1282026.07 | 46.3 | |
| Chat-UniVi-V1.5LLM Size=7B, #Frames=642026.07 | 40.6 | |
| Video-LLaVALLM Size=7B, #Frames=82026.07 | 39.9 | |
| ShareGPT4VideoLLM Size=8B, #Frames=162026.07 | 39.9 | |
| VideoChat2LLM Size=7B, #Frames=162026.07 | 39.5 |