Model,Model Type,Model Size,Frames,Avg. All,Avg. Action,Avg. Direction,Avg. Speed,Avg. Event Order,Avg. Attribute Change,Avg. Multi-Choice,Avg. Yes/No,Avg. Caption Matching,Avg. Caption Generation,Action. Multi-Choice,Action. Yes/No,Action. Caption Matching,Action. Caption Generation,Direction. Multi-Choice,Direction. Yes/No,Direction. Caption Matching,Direction. Caption Generation,Speed. Multi-Choice,Speed. Yes/No,Speed. Caption Matching,Speed. Caption Generation,Event Order. Multi-Choice,Event Order. Yes/No,Event Order. Caption Matching,Event Order. Caption Generation,Attribute Change. Multi-Choice,Attribute Change. Yes/No,Attribute Change. Caption Matching,Attribute Change. Caption Generation,Notes
[Video-LLaVA-7B](https://huggingface.co/LanguageBind/Video-LLaVA-7B),VideoLLM,7B,8,49.77,71.41,42.4,41.94,45.66,45.74,45.57,56.38,63.34,34.83,76.04,74.32,87.88,50.76,35.22,51.82,53.82,28.67,35.65,50.28,58.42,23.2,37.75,49.21,59.0,38.25,40.97,51.12,58.33,33.59,
[VideoChat2-vicuna-stage3](https://github.com/OpenGVLab/Ask-Anything/blob/main/video_chat2/scripts/videochat_vicuna/config_7b_stage3.py),VideoLLM,7B,16,48.81,69.19,42.02,43.44,42.2,45.31,42.91,58.01,53.69,38.52,76.92,72.8,73.4,54.04,29.55,53.85,51.68,30.96,33.44,53.82,49.48,32.73,35.43,51.31,48.0,34.25,36.81,53.79,45.83,41.41,
[LLaMA-VID-7B-short-video](https://huggingface.co/YanweiLi/llama-vid-7b-full-224-video-fps-1),VideoLLM,7B,1fps,45.61,62.11,39.45,38.62,42.63,44.11,38.04,52.96,56.02,34.78,61.24,63.01,73.4,53.03,29.85,48.79,50.46,27.98,29.65,49.16,51.2,21.91,33.77,48.43,53.67,35.5,34.03,52.68,51.74,35.94,
[mPLUG-Owl-video](https://huggingface.co/MAGAer13/mplug-owl-llama-7b-video),VideoLLM,7B,8,44.15,56.93,38.82,40.44,39.74,43.82,36.39,54.42,48.5,34.43,55.92,64.36,57.24,46.46,32.24,50.61,41.9,28.21,31.55,51.21,43.64,30.41,27.48,51.31,48.67,31.25,32.99,52.01,51.74,36.46,
[PandaGPT-13B-maxlen-400](https://huggingface.co/openllmplayground/pandagpt_13b_max_len_400),VideoLLM,13B,10,41.64,41.77,41.08,40.77,42.2,42.54,34.37,51.81,51.56,27.5,36.69,53.04,49.16,23.74,34.03,49.6,55.96,25.69,33.75,50.84,49.48,26.03,32.12,53.66,54.33,29.75,35.07,52.23,48.26,32.55,
[Valley2-7B](https://huggingface.co/luoruipu1/Valley2-7b),VideoLLM,7B,8,37.49,43.01,34.86,36.07,36.05,37.07,29.56,53.49,34.6,26.35,43.49,58.11,36.7,24.75,29.55,52.02,33.64,20.41,24.92,52.51,36.77,21.91,18.54,50.26,36.0,35.75,29.86,52.9,29.86,29.43,
[Video-ChatGPT](https://github.com/mbzuai-oryx/Video-ChatGPT),VideoLLM,7B,100,43.54,54.59,39.51,38.55,42.56,41.76,37.66,50.67,53.69,31.84,61.54,52.53,69.02,40.91,28.96,50.0,49.24,28.44,29.02,49.53,47.42,24.48,36.09,51.05,52.67,31.75,30.9,50.0,50.35,33.85,
[Video-LLaMA-2-13B](https://huggingface.co/DAMO-NLP-SG/Video-LLaMA-2-13B-Finetuned),VideoLLM,13B,8,43.4,61.86,34.67,37.05,42.56,39.7,31.27,53.73,54.16,32.24,52.07,68.07,70.71,54.29,25.07,45.95,45.26,21.33,28.08,48.79,56.01,13.92,26.82,51.83,52.0,38.5,22.22,50.89,47.57,33.85,
Gemini-1.5-pro,VideoLLM,-,1fps,67.1,91.68,48.99,55.9,64.67,73.79,63.92,70.32,77.45,57.88,96.75,88.18,97.98,87.88,43.28,54.45,63.0,36.7,55.21,63.87,65.64,38.14,55.63,68.85,81.0,55.25,67.71,73.21,80.9,73.7,
[LLaVA-1.5-13B](https://huggingface.co/liuhaotian/llava-v1.5-13b),ImageLLM,13B,1,51.3,78.43,42.21,42.86,43.71,46.95,47.41,56.38,64.27,38.42,85.5,74.66,92.59,67.42,31.04,48.79,57.49,31.88,40.06,48.98,58.76,24.74,35.43,49.48,59.0,33.0,42.36,55.36,53.82,35.42,
[SPHINX-v2](https://huggingface.co/Alpha-VLLM/LLaMA2-Accessory/tree/main/finetune/mm/SPHINX/SPHINX-v2-1k),ImageLLM,13B,1,51.92,81.21,40.33,43.64,46.89,45.24,50.63,59.07,64.34,34.88,89.94,79.05,93.27,67.93,36.72,51.21,55.96,19.04,43.53,54.75,54.3,20.36,37.42,54.45,59.67,37.25,42.36,50.45,59.03,30.99,
[Qwen-VL-Chat](https://huggingface.co/Qwen/Qwen-VL-Chat),ImageLLM,7B,1,52.9,80.16,42.53,47.42,46.89,45.1,51.08,60.01,64.07,37.28,88.46,81.42,91.58,62.63,37.01,51.62,54.13,27.75,41.32,59.78,54.98,29.64,41.39,50.79,63.67,34.75,44.44,49.11,56.6,32.29,
[Reka-Core-20240501](https://docs.reka.ai/image-video-and-audio-chat),VideoLLM,-,-,64.31,93.22,46.92,53.23,63.58,63.42,61.08,65.14,74.05,58.53,96.15,90.71,97.31,91.41,40.9,52.83,58.1,36.47,49.21,54.0,63.92,47.42,57.28,57.33,81.0,61.25,60.42,64.96,71.18,58.07,
[LLaVA-NeXT-Video-32B-Qwen](https://huggingface.co/lmms-lab/LLaVA-NeXT-Video-32B-Qwen),VideoLLM,32B,32,66.74,90.2,46.04,49.25,75.94,73.08,68.67,69.38,76.51,54.64,95.86,86.82,96.3,85.86,42.09,54.05,59.02,30.28,49.53,57.73,62.54,27.32,78.48,74.35,89.0,65.75,78.47,72.99,77.08,66.15,
[LLaVA-NeXT-Video-7B-DPO](https://huggingface.co/lmms-lab/LLaVA-NeXT-Video-7B-DPO),VideoLLM,7B,32,53.75,81.21,43.72,45.66,47.4,48.51,50.57,61.19,63.01,40.22,87.57,83.78,92.59,63.38,35.82,51.42,54.74,32.8,41.32,56.42,55.33,27.06,39.74,54.97,56.0,39.5,45.83,53.12,56.94,38.8,
[LongVA-7B](https://huggingface.co/lmms-lab/LongVA-7B),VideoLLM,7B,32,56.95,86.44,44.22,45.79,53.03,53.34,56.14,62.13,65.67,44.71,92.31,86.15,94.61,75.76,36.72,50.4,54.43,35.32,43.22,53.07,53.26,32.22,54.3,61.78,64.33,35.25,52.43,54.46,62.5,45.83,
[Llama-3-VILA1.5-8B](https://huggingface.co/Efficient-Large-Model/Llama-3-VILA1.5-8B),VideoLLM,8B,8,58.79,85.89,45.29,47.68,55.64,58.03,56.39,63.64,68.93,47.16,92.9,84.8,94.95,74.75,33.73,52.23,58.72,36.24,44.16,54.38,60.48,31.7,50.0,61.26,66.0,46.75,60.07,61.38,65.28,47.14,
[InternVL2-8B](https://huggingface.co/OpenGVLab/InternVL2-8B),VideoLLM,8B,8,66.03,88.79,48.68,52.64,69.94,70.17,65.57,68.24,77.11,55.39,93.79,84.8,96.63,84.6,43.88,53.24,59.94,38.76,51.1,61.27,67.01,31.19,67.22,70.68,84.0,60.75,71.88,69.2,79.51,63.02,
[InternLM-XComposer-2.5](https://huggingface.co/internlm/internlm-xcomposer2d5-7b),VideoLLM,7B,32,62.1,89.96,46.04,49.45,60.55,63.42,61.33,64.49,72.39,52.05,94.97,84.29,97.98,88.13,42.69,52.02,57.19,33.49,41.64,56.24,60.48,38.14,59.93,64.14,75.33,46.5,66.67,62.28,72.22,55.73,
[LLaVA-OneVision-Qwen-2-7B](https://huggingface.co/lmms-lab/llava-onevision-qwen2-7b-ov),VideoLLM,7B,32,64.22,88.35,45.79,47.62,70.23,69.39,64.81,69.67,73.79,49.9,96.45,85.98,95.96,79.29,40.6,55.26,56.88,30.73,45.43,57.36,61.86,25.26,69.54,76.18,81.33,56.75,72.22,73.21,74.31,59.11,
[MiniCPM-V-2.6](https://huggingface.co/openbmb/MiniCPM-V-2_6),VideoLLM,8B,64,66.26,92.36,49.37,52.25,68.5,68.32,63.04,68.49,76.38,58.48,96.75,87.67,98.99,90.66,42.09,55.67,60.55,39.45,45.11,58.29,64.26,40.72,59.6,65.97,86.33,64.25,71.18,71.65,72.92,58.85,
[GPT-4o](https://openai.com/index/hello-gpt-4o/),VideoLLM,-,8,73.75,95.75,57.04,58.9,80.49,76.85,70.95,73.66,80.84,70.76,98.22,91.89,99.33,96.72,52.84,61.13,70.03,45.87,52.05,60.15,67.7,56.19,73.18,79.06,88.0,81.75,78.47,75.0,79.86,75.52,
[Qwen2-VL-7B-Instruct](https://huggingface.co/Qwen/Qwen2-VL-7B-Instruct),VideoLLM,7B,1fps,67.85,91.87,54.9,54.73,66.33,70.6,68.48,72.77,77.31,54.24,97.63,87.67,97.98,88.64,51.34,62.75,65.75,40.6,51.42,63.87,62.54,38.92,67.88,71.99,84.0,46.5,73.61,75.45,77.08,57.81,max_pixels: 720 * 720