OpenCodePapers

zero-shot-video-question-answer-on-longvideobench

Video Question AnsweringZero-Shot Video Question Answer
Dataset Link
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
PaperCodeAccuracy (% )ModelNameReleaseDate
Molmo2: Open Weights and Data for Vision-Language Models with Video Understanding and Grounding✓ Link68.0Molmo2-4B2026-01-15
Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context✓ Link66.7Gemini 1.5 Pro2024-03-08
InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models✓ Link65.7InternVL3-78B2025-04-14
Video-RAG: Visually-aligned Retrieval-Augmented Long Video Comprehension✓ Link65.4Video-RAG (based on LLaVA-Video)2024-11-20
GPT-4o: Visual perception performance of multimodal large language models in piglet activity understanding64.0GPT-4o2024-06-14
MiniCPM-V 4.5: Cooking Efficient MLLMs via Architecture, Data, and Training Recipe✓ Link63.9MiniCPM-V 4.52025-09-16
Video Instruction Tuning With Synthetic Data61.9LLaVA-Video2024-10-03
LLaVA-OneVision: Easy Visual Task Transfer✓ Link61.3LLaVA-OneVision-72B2024-08-06
Qwen2.5-VL Technical Report✓ Link60.7Qwen2.5-VL-72B2025-02-19
TimeViper: A Hybrid Mamba-Transformer Vision-Language Model for Efficient Long Video Understanding✓ Link54.1TimeViper2025-11-20