whatmodel.sh / video / CogVideoX-2B

CogVideoX-2B

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

CogVideoX-2B is an open-weights video model from Zhipu AI (THUDM), released August 2024. Its published size is 2B parameters. Licence: apache-2.0.

36.1±2.1#77 of 93 video modelsfrom 4 benchmark results across 3 sources

$ how it compares

The video models ranked either side of CogVideoX-2B, out of 139 with a score, then more from Zhipu AI (THUDM).

#modelscorewhereRAMreleased
112LTX-2.3 Lightricks36±3api $0.040/sec +local~34GB2026-03
113CogVideoX1.5-5B Zhipu AI (THUDM)36±2local~38GB2024-11
114CogVideoX-2B Zhipu AI (THUDM)36±2local~5.3GB2024-08
115Pika 2.2 Pika36±2api $0.040/sec2025-02
116Allegro Rhymes AI35±2local~9.3GB2024-10
more from Zhipu AI (THUDM)
85CogVideoX-5B-I2V Zhipu AI (THUDM)46±4local~13GB2024-09
91CogVideoX-5B Zhipu AI (THUDM)44±2api price unknown +local~13GB2024-08

see all video models · find the ones that fit your machine

$ what it's good at

text to video▆▄▃▄▆▅█▃▂▂▂▄ #52 of 68 · 98% of its score · 4 tests

Beats CogVideoX1.5-5B here, though they rank higher overall. Trails Pika 2.2 here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
GenAI-Arena text-to-video (anonymous Elo)1053.5 ✎▃▃···▅▃·▃▅·█ beats 6/10avg 1229▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 42%sharp
VideoPhy (human, SA=1 & PC=1)18.6% ✎▄▄▄·▄▄█····▄ beats 4/7medium · avg 48%▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 37%sharp
VideoPhy (auto-evaluator average)39.0% ✎▅·▃▅··▃█···▃ beats 5/9medium · avg 56%▇▇▇▇ 11%at its level
ChronoMagic-Bench-150 GPT4o-MTScore (time-lapse metamorphosis)61.8% ✎▄·▄█▄▄▄·▄·▄█ beats 9/10medium · avg 71%▇▇▇ 8.5%at its level

$ where to run it

Other ways to run it

  1. Run it as published · needs ~5.3 GB, NVIDIA only · from huggingface.co/THUDM (official)
    python -c "import torch;from diffusers import CogVideoXPipeline;from diffusers.utils import export_to_video;p=CogVideoXPipeline.from_pretrained('THUDM/CogVideoX-2b',torch_dtype=torch.float16);p.enable_sequential_cpu_offload();p.vae.enable_slicing();p.vae.enable_tiling();export_to_video(p(prompt='A panda playing a tiny guitar in a bamboo forest',num_inference_steps=50,num_frames=49,guidance_scale=6).frames[0],'output.mp4',fps=8)"

Weights and documentation: huggingface.co/zai-org ↗

$ how the score was worked out · VideoPhy / VideoPhy-2 officia… 48%, GenAI-Arena (TIGER-Lab) text-… 42%, ChronoMagic-Bench: A Benchmar… 8%

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
VideoPhy / VideoPhy-2 official repository leaderboards2▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 48%MIT
GenAI-Arena (TIGER-Lab) text-to-video leaderboard1▇▇▇▇▇▇▇▇▇▇▇▇▇ 42%MIT
ChronoMagic-Bench: A Benchmark for Metamorphic Evaluation of Text-to-Time-lapse Video Generation1▇▇▇ 8%CC-BY-4.0
prior · pull toward the middle▇ 2%
weighted average= 36.1

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json