# Vchitect/VBench

[CVPR2024 Highlight] VBench - We Evaluate Video Generation

Repository: https://github.com/Vchitect/VBench
Canonical: https://ross.abutalabs.com/products/vbench
Homepage: https://vchitect.github.io/VBench-project/
Language: Python
License: Apache-2.0
License Family: permissive
Topics: aigc, evaluation-kit, gen-ai, stable-diffusion, text-to-video, video-generation, benchmark, dataset
Last push: 2026-08-21T15:28:08+00:00

## Health v2 (maintenance only)
Score: 61/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 98, release rhythm 8, longevity 72
- inputs: {"age_days": 1010, "days_push": 12, "days_rel": 729, "gap_med": null, "n_releases_24m": 1}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1747, forks 132 (observed 2026-08-28T04:05:30.869405+00:00)

## What it is
VBench is a comprehensive benchmark suite for evaluating video generative models across hierarchical evaluation dimensions, with prompt suites, sampled videos, and automatic evaluation methods. It supports text-to-video and image-to-video evaluation and aligns with human preference annotations.

## Use cases
- benchmark my text-to-video model
- compare video generation models on quality dimensions
- evaluate image-to-video generation
- find a leaderboard for video generation models
- measure trustworthiness of generative video models
- get automatic metrics for generated videos

## When to choose
- you need fine-grained, objective evaluation of video generative models
- you want results aligned with human preference annotations
- you need a standardized prompt suite and evaluation pipeline for reproducible comparisons

## When to avoid
- you need to generate videos rather than evaluate them
- your task is image-only generation evaluation
- you need a lightweight metric without GPU-heavy pipelines

## Facets
- artifact type: library
- maturity: active
- function: benchmarking, machine-learning
- domain: artificial-intelligence, machine-learning
- platform: python
- tags: text-to-video, video-generation, aigc, benchmark-suite, evaluation-kit, stable-diffusion, evaluation, video, gpu

## Member repositories
- Vchitect/VBench (main) score 61

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:30.869405+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:29:01.989120+00:00, confidence not recorded.
  - readme: https://github.com/Vchitect/VBench (fetched 2026-08-28T04:05:30.869405+00:00, sha 76b1eae4db74)
  - homepage: https://vchitect.github.io/VBench-project/ (fetched 2026-08-29T11:06:50.517437+00:00, sha c6ea5404de37)
  - registry_pypi: https://pypi.org/pypi/vbench/json (fetched 2026-08-29T11:06:50.520308+00:00, sha 1e2b8a4563b5)
- Data as of 2026-08-30T08:39:29.467469+00:00.