LibreAI / Video / t2v_metrics
// video Β· linzhiqiu/t2v_metrics

t2v_metrics

Evaluate text-to-video models with VQAScore-based metrics.

πŸ’Ž Hidden gemβ˜… 603 starsβ‘‚ 75 forksApache-2.0updated 2026-06
β–¦ visual β€” this project's README ships no embeddable screenshot,
so here's a placeholder. The guide below is complete regardless.
// project facts

Project facts

t2v_metrics is an open-source video project by linzhiqiu (linzhiqiu/t2v_metrics on GitHub). It has 603 GitHub stars and 75 forks, is released under the Apache-2.0 license, and was last updated on GitHub on 5 June 2026.

GitHub stars
603
Forks
75
License
Apache-2.0
Last updated on GitHub
5 June 2026
Category
Video
Listing
Hidden gem β€” a shorter page for a lesser-known project
Official site
Official site
// how it works

How it works

  1. Install the toolkit

    Set up the evaluation code.

  2. Prepare model outputs

    Gather generated images, videos, or 3D.

  3. Run VQAScore

    Score prompt alignment across your samples.

  4. Compare and report

    Benchmark against other models fairly.

// features

Why people love it

VQAScore metric

Principled prompt-alignment scoring.

Three modalities

Image, video, and 3D evaluation.

Published method

Backed by an academic paper.

Model comparison

Fair numbers instead of cherry-picks.

Research tool

Built for rigorous evaluation.

Apache 2.0

Open for the community.

Advertisement
// run it

Run it yourself

  1. Step 1 β€” Clone the repo

    Download the source and move into the project folder.

    git clone https://github.com/linzhiqiu/t2v_metrics.git
    cd t2v_metrics
  2. Step 2 β€” Check the README

    Every project documents its own setup; start with install requirements.

    cat README.md | head -60
  3. Step 3 β€” Run it

    Follow the project's launch command β€” Docker, pip, npm, or a binary release.

    # see README for the exact launch command
// use cases

What you can build

Advertisement
// faq

Questions, answered

Is t2v_metrics free to use?

Apache 2.0 licensed β€” free for research and commercial use.

What is VQAScore?

A published metric for visual-question-answering-based alignment scoring.

Does it need GPUs?

Scoring models need compute β€” check the README.

Can it evaluate my custom model?

Yes β€” point it at any model's outputs.

Why not just look at samples?

Human eyeballing does not scale; metrics do.

// keep exploring

Similar tools

Video

HandBrake

HandBrake's development repository

β˜… 24.6k starsCustom
View guide β†’
Video

awesome-generative-ai-apps

50+ open-source generative AI apps you can clone, deploy, and monetize β€” image generators, video tools, virtual try-ons, AI SaaS templates, and platform inte…

β˜… 3.4k starsMIT
View guide β†’
Video

Awesome-World-Models

A comprehensive list of papers for the definition of World Models and using World Models for General Video Generation, Embodied AI, and Autonomous Driving, i…

β˜… 2.0k starsBSD-3-Clause
View guide β†’
Video

CutScript

Edit video by editing text β€” an AI-powered text-based video editor.

β˜… 262 starsMIT
View guide β†’
Video

synctoon

Python-based automated 2D animation utility that produces videos from text scripts

β˜… 50 starsGPL-3.0
View guide β†’
Video

Seedance-2.5-API

Python wrapper for Seedance 2.5 β€” T2V, image-to-video and 4K human faces.

β˜… 488 starsMIT
View guide β†’
// alternatives

Alternatives in Video

The most-starred other Video projects in the LibreAI catalogue β€” the realistic shortlist if t2v_metrics is not the right fit.

Get t2v_metrics

Free and open β€” grab the source, star the repo, and start creating.