Metric: Acc@GQA (higher is better)
| # | Model↕ | Acc@GQA▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | DeVi (Gemini 2.0) | 28.9 | No | Question-Answering Dense Video Events | 2024-09-06 | Code |
| 2 | VideoMind(7B) | 28.2 | No | VideoMind: A Chain-of-LoRA Agent for Long Video ... | 2025-03-17 | Code |
| 3 | DeVi (GPT-4) | 28 | No | Question-Answering Dense Video Events | 2024-09-06 | Code |
| 4 | LLoVi (GPT-4) | 26.8 | No | A Simple LLM Framework for Long-Range Video Ques... | 2023-12-28 | Code |
| 5 | VideoMind (2B) | 25.2 | No | VideoMind: A Chain-of-LoRA Agent for Long Video ... | 2025-03-17 | Code |
| 6 | VideoStreaming | 17.8 | No | Streaming Long Video Understanding with Large La... | 2024-05-25 | - |
| 7 | LangRepo (12B) | 17.1 | No | Language Repository for Long Video Understanding | 2024-03-21 | Code |
| 8 | LLoVi (7B) | 11.2 | No | A Simple LLM Framework for Long-Range Video Ques... | 2023-12-28 | Code |
| 9 | Mistral (7B) | 9.2 | No | Mistral 7B | 2023-10-10 | Code |