Image & vision

PhyGenBench

by OpenGVLab

What is PhyGenBench? PhyGenBench is an image & vision eval made by OpenGVLab. An ICML 2025 benchmark for physical commonsense in video generation ('Towards World Simulator'). Code and data are on GitHub. The official repo had no LICENSE file and no last code push later than 25 Oct 2024 when checked on 14 Nov 2026, so treat it as a documented earlier physics suite rather than a currently maintained leaderboard, and leave the licence empty.

Page last updated 2026-10-18 · Evals Review

Editor note

An ICML 2025 benchmark for physical commonsense in video generation ('Towards World Simulator'). Code and data are on GitHub. The official repo had no LICENSE file and no last code push later than 25 Oct 2024 when checked on 14 Nov 2026, so treat it as a documented earlier physics suite rather than a currently maintained leaderboard, and leave the licence empty.

What it tests
Physical commonsense in generated video.
Original source
github.com/OpenGVLab/PhyGenBench
Last updated
Repo last push 2024-10-25 (GitHub API, earlier read 14 Nov 2026); LICENSE still absent 14 Nov 2026
License
Not checked yet — see source

Facts come only from the linked source (checked 15–16 Oct 2026). Credit belongs to the creators. We don't score or rank evals: how we review.

Reviews

No reviews yet.

Reviews are written by people who sign in with GitHub, one per account per eval. We can't confirm that each reviewer used the eval; we remove reviews that break our rules, never just for being negative, and we never add, buy or edit reviews. Averages, when shown, come only from these reviews.

Sign in with GitHub to review

Related image & vision evals

  • MMMU — Expert-level multimodal understanding and reasoning.
  • MathVista — Mathematical reasoning over figures, plots and images.
  • MMBench — Broad perception and reasoning skills of vision-language models.
  • ChartQA — Question answering over charts.

All Image & vision evals · Comparison table

Discussion

No one has discussed PhyGenBench yet. Have you run it? Share what it caught, what it missed or how you set it up.

Sign in with GitHub to comment

Be specific and kind. No promotion or affiliate links. Community rules · All discussions