Vchitect/VBench

This repository provides unified implementations for the VBench series of works, supporting comprehensive evaluation of video generative models across a wide spectrum of capabilities and settings.

If your questions are not addressed in this README, please contact Ziqi Huang at ZIQI002 [at] e [dot] ntu [dot] edu [dot] sg.

Table of Contents

• Overview - See this section for component locations and the differences between VBench, VBench++, and VBench-2.0.
• Updates
• Evaluation Results
• Video Generation Models Info
• Installation
• Usage
• Prompt Suite
• Sampled Videos
• Evaluation Method Suite
• Citation and Acknowledgement

:mega: Overview

This repository provides unified implementations for the VBench series of works, supporting comprehensive evaluation of video generative models across a wide spectrum of capabilities and settings.

(1) VBench

TL;DR: Evaluating Video Generation — Benchmark • Evaluation Dimensions • Evaluation Methods • Human Alignment • Insights

VBench: Comprehensive Benchmark Suite for Video Generative Models

Ziqi Huang^∗, Yinan He^∗, Jiashuo Yu^∗, Fan Zhang^∗, Chenyang Si, Yuming Jiang, Yuanhan Zhang, Tianxing Wu, Qingyang Jin, Nattapol Chanpaisit, Yaohui Wang, Xinyuan Chen, Limin Wang, Dahua Lin⁺, Yu Qiao⁺, Ziwei Liu⁺

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024

We propose VBench, a comprehensive benchmark suite for video generative models. We design a comprehensive and hierarchical Evaluation Dimension Suite to decompose "video generation quality" into multiple well-defined dimensions to facilitate fine-grained and objective evaluation. For each dimension and each content category, we carefully design a Prompt Suite as test cases, and sample Generated Videos from a set of video generation models. For each evaluation dimension, we specifically design an Evaluation Method Suite, which uses carefully crafted method or designated pipeline for automatic objective evaluation. We also conduct Human Preference Annotation for the generated videos for each dimension, and show that VBench evaluation results are well aligned with human perceptions. VBench can provide valuable insights from multiple perspectives.

Note: The code and README for the VBench components are located here, relative path: ..

@InProceedings{huang2023vbench,
    title={{VBench}: Comprehensive Benchmark Suite for Video Generative Models},
    author={Huang, Ziqi and He, Yinan and Yu, Jiashuo and Zhang, Fan and Si, Chenyang and Jiang, Yuming and Zhang, Yuanhan and Wu, Tianxing and Jin, Qingyang and Chanpaisit, Nattapol and Wang, Yaohui and Chen, Xinyuan and Wang, Limin and Lin, Dahua and Qiao, Yu and Liu, Ziwei},
    booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
    year={2024}
}

(2) VBench++

TL;DR: Extends VBench with (1) VBench-I2V for image-to-video, (2) VBench-Long for long videos, and (3) VBench-Trustworthiness covering fairness, bias, and safety.

VBench++: Comprehensive and Versatile Benchmark Suite for Video Generative Models

Ziqi Huang^∗, Fan Zhang^∗, Xiaojie Xu, Yinan He, Jiashuo Yu, Ziyue Dong, Qianli Ma, Nattapol Chanpaisit, Chenyang Si, Yuming Jiang, Yaohui Wang, Xinyuan Chen, Ying-Cong Chen, Limin Wang, Dahua Lin⁺, Yu Qiao⁺, Ziwei Liu⁺

IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2025

VBench++ supports a wide range of video generation tasks, including text-to-video and image-to-video, with an adaptive Image Suite for fair evaluation across different settings. It evaluates not only technical quality but also the trustworthiness of generative models, offering a comprehensive view of model performance. We continually incorporate more video generative models into VBench to inform the community about the evolving landscape of video generation.

Note: The code and README for the VBench++ components are located at:

• (1) VBench-I2V (image-to-video): link, relative path: vbench2_beta_i2v
• (2) VBench-Long (long video evaluation): link, relative path: vbench2_beta_long
• (3) VBench-Trustworthiness (fairness, bias, and safety): link, relative path: vbench2_beta_trustworthiness

*These modules belong to VBench++, not VBench, or VBench-2.0. However, to maintain backward compatibility for users who have already installed the repository, we preserve the original relative path names and provide this clarification here.

• @article{huang2025vbench++,

title={{VBench++}: Comprehensive and Versatile Benchmark Suite for Video Generative Models}, author={Huang, Ziqi and Zhang, Fan and Xu, Xiaojie and He, Yinan and Yu, Jiashuo and Dong, Ziyue and Ma, Qianli and Chanpaisit, Nattapol and Si, Chenyang and Jiang, Yuming and Wang, Yaohui and Chen, Xinyuan and Chen, Ying-Cong and Wang, Limin and Lin, Dahua and Qiao, Yu and Liu, Ziwei}, journal={IEEE Transactions on Pattern Analysis and Machine Intelligence}, year={2025}, doi={10.1109/TPAMI.2025.3633890} }

(3) VBench-2.0

TL;DR: Extends VBench to evaluate intrinsic faithfulness — a key challenge for next-generation video generation models.

VBench-2.0: Advancing Video Generation Benchmark Suite for Intrinsic Faithfulness

Dian Zheng^∗, Ziqi Huang^∗, Hongbo Liu, Kai Zou, Yinan He, Fan Zhang, Yuanhan Zhang, Jingwen He, Wei-Shi Zheng⁺, Yu Qiao⁺, Ziwei Liu⁺

Overview of VBench-2.0. (a) Scope of VBench-2.0. Video generative models have progressed from achieving superficial faithfulness in fundamental technical aspects such as pixel fidelity and basic prompt adherence, to addressing more complex challenges associated with intrinsic faithfulness, including commonsense reasoning, physics-based realism, human motion, and creative composition. While VBench primarily assessed early-stage technical quality, VBench-2.0 expands the benchmarking framework to evaluate these advanced capabilities, ensuring a more comprehensive assessment of next-generation models. (b) Evaluation Dimension of VBench-2.0. VBench-2.0 introduces a structured evaluation suite comprising five broad categories and 18 fine-grained capability dimensions.

Note: The code and README for the VBench-2.0 components are located at link, relative path: VBench-2.0.

@article{zheng2025vbench2,
    title={{VBench-2.0}: Advancing Video Generation Benchmark Suite for Intrinsic Faithfulness},
    author={Zheng, Dian and Huang, Ziqi and Liu, Hongbo and Zou, Kai and He, Yinan and Zhang, Fan and Zhang, Yuanhan and He, Jingwen and Zheng, Wei-Shi and Qiao, Yu and Liu, Ziwei},
    journal={arXiv preprint arXiv:2503.21755},
    year={2025}
}

:fire: Updates

• [03/2026] VBench-I2V Arena released: View the generated videos here, and vote for your preferred video. You can explore videos generated by your chosen models following your chosen text prompts.
• [11/2025] VBench++ accepted to TPAMI:
• [05/2025] We support evaluating customized videos for VBench-2.0! See here for instructions.
• [04/2025] Human Anomaly Detection for AIGC Videos: We release the pipeline for evaluating human anatomical quality in AIGC videos, including a manually human anomaly dataset on real and AIGC videos, and the training pipeline for anomaly detection.
• [03/2025] :fire: Major Update! We released VBench-2.0! :fire: Video generative models have progressed from achieving superficial faithfulness in fundamental technical aspects such as pixel fidelity and basic prompt adherence, to addressing more complex challenges associated with intrinsic faithfulness, including commonsense reasoning, physics-based realism, human motion, and creative composition. While VBench primarily assessed early-stage technical quality, VBench-2.0 expands the benchmarking framework to evaluate these advanced capabilities, ensuring a more comprehensive assessment of next-generation models.
• [01/2025] PyPI Updates: v0.1.5 preprocessing bug fixes, torch>=2.0 support.
• [01/2025] VBench Arena released: View the generated videos here, and vote for your preferred video. This demo features over 180,000 generated videos, and you can explore videos generated by your chosen models (we already support 40 models) following your chosen text prompts.
• [09/2024] VBench-Long Leaderboard available: Our VBench-Long leaderboard now has 10 long video generation models. VBench leaderboard now has 40 text-to-video (both long and short) models. All video generative models are encouraged to participate!
• [09/2024] PyPI Updates: PyPI package is updated to version 0.1.4: bug fixes and multi-gpu inference.
• [08/2024] Longer and More Descriptive Prompts: Available Here! We follow CogVideoX's prompt optimization technique to enhance VBench prompts using GPT-4o, making them longer and more descriptive without altering their original meaning.
• [08/2024] VBench Leaderboard update: Our leaderboard has 28 T2V models, 12 I2V models so far. All video generative models are encouraged to participate!
• [06/2024] :fire: VBench-Long :fire: is ready to use for evaluating longer Sora-like videos!
• [06/2024] Model Info Documentation: Information on video generative models in our VBench Leaderboard

is documented HERE.

• [05/2024] PyPI Update: PyPI package vbench is updated to version 0.1.2. This includes changes in the preprocessing for high-resolution images/videos for imaging_quality, support for evaluating customized videos, and minor bug fixes.
• [04/2024] We release all the videos we sampled and used for VBench evaluation. See details here.
• [03/2024] :fire: VBench-Trustworthiness :fire: We now support evaluating the trustworthiness (e.g., culture, fairness, bias, safety) of video generative models.
• [03/2024] :fire: VBench-I2V :fire: We now support evaluating Image-to-Video (I2V) models. We also provide Image Suite.
• [03/2024] We support evaluating customized videos! See here for instructions.
• [02/2024] VBench accepted to CVPR 2024 as Highlight:
• [01/2024] PyPI package is released! . Simply pip install vbench.
• [12/2023] :fire: VBench :fire: Evaluation code released for 16 Text-to-Video (T2V) evaluation dimensions.
• ['subject_consistency', 'background_consistency', 'temporal_flickering', 'motion_smoothness', 'dynamic_degree', 'aesthetic_quality', 'imaging_quality', 'object_class', 'multiple_objects', 'human_action', 'color', 'spatial_relationship', 'scene', 'temporal_style', 'appearance_style', 'overall_consistency']
• [11/2023] Prompt Suites released. (See prompt lists here)

:mortar_board: Evaluation Results

See our leaderboard for the most updated ranking and numerical results (with models like Gen-3, Kling, Pika).

We visualize the evaluation results of the 12 most recent top-performing long video generation models across 16 VBench dimensions.

Additionally, we present radar charts separately for the evaluation results of open-source and closed-source models. The results are normalized per dimension for clearer comparisons.

:trophy: Leaderboard

See numeric values at our Leaderboard :1st_place_medal::2nd_place_medal::3rd_place_medal:

:film_projector: Model Info

See model info for video generation models we used for evaluation.

:hammer: Installation

Install with pip

pip install torch torchvision --index-url https://download.pytorch.org/whl/cu118 # or any other PyTorch version with CUDA<=12.1
pip install vbench

To evaluate some video generation ability aspects, you need to install detectron2 via:

   pip install detectron2@git+https://github.com/facebookresearch/detectron2.git
   

If there is an error during detectron2 installation, see here. Detectron2 is working only with CUDA 12.1 or 11.X.

Download VBench_full_info.json to your running directory to read the benchmark prompt suites.

Install with git clone

git clone https://github.com/Vchitect/VBench.git pip install torch torchvision --index-url https://download.pytorch.org/whl/cu118 # or other version with CUDA<=12.1 pip install VBench If there is an error during detectron2 installation, see here.

Usage

Use VBench to evaluate videos, and video generative models.

• A Side Note: VBench is designed for evaluating different models on a standard benchmark. Therefore, by default, we enforce evaluation on the standard VBench prompt lists to ensure fair comparisons among different video generation models. That's also why we give warnings when a required video is not found. This is done via defining the set of prompts in VBench_full_info.json. However, we understand that many users would like to use VBench to evaluate their own videos, or videos generated from prompts that does not belong to the VBench Prompt Suite, so we also added the function of Evaluating Your Own Videos. Simply set mode=custom_input, and you can evaluate your own videos.

[New] Evaluate Your Own Videos

We support evaluating any video. Simply provide the path to the video file, or the path to the folder that contains your videos. There is no requirement on the videos' names.

• Note: We support customized videos / prompts for the following dimensions: 'subject_consistency', 'background_consistency', 'motion_smoothness', 'dynamic_degree', 'aesthetic_quality', 'imaging_quality'

To evaluate videos with customized input prompt, run our script with --mode=custom_input:

python evaluate.py \
    --dimension $DIMENSION \
    --videos_path /path/to/folder_or_video/ \
    --mode=custom_input

alternatively you can use our command:

vbench evaluate \
    --dimension $DIMENSION \
    --videos_path /path/to/folder_or_video/ \
    --mode=custom_input

To evaluate using multiple gpus, we can use the following commands:

torchrun --nproc_per_node=${GPUS} --standalone evaluate.py ...args...

or

vbench evaluate --ngpus=${GPUS} ...args...

Evaluation on the Standard Prompt Suite of VBench

Command Line

vbench evaluate --videos_path $VIDEO_PATH --dimension $DIMENSION

For example:

vbench evaluate --videos_path "sampled_videos/lavie/human_action" --dimension "human_action"

Python

from vbench import VBench
my_VBench = VBench(device, , )
my_VBench.evaluate(
    videos_path = <video_path>,
    name = ,
    dimension_list = [, , ...],
)

For example:

from vbench import VBench
my_VBench = VBench(device, "vbench/VBench_full_info.json", "evaluation_results")
my_VBench.evaluate(
    videos_path = "sampled_videos/lavie/human_action",
    name = "lavie_human_action",
    dimension_list = ["human_action"],
)

Evaluation of Different Content Categories

command line

vbench evaluate \
    --videos_path $VIDEO_PATH \
    --dimension $DIMENSION \
    --mode=vbench_category \
    --category=$CATEGORY

or ``` python evaluate.py \ --dimension $DIMENSION \ --videos_path /path/to/folder_