whatmodel.sh / video / Show-1

Show-1

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

Show-1 is an open-weights video model from Show Lab / NUS, released October 2023. Its published size is 6B parameters. Licence: cc-by-nc-4.0.

Catalogue identity notes

Original Show-1 multi-stage text-to-video pipeline, 576×320, 8 FPS; not a single component checkpoint.

Model information sources

· Weights licence ↗ · Parameter count ↗

6B = total over the four Show-1 stages as the makers tabulate them (keyframes 1.7B, temporal interpolation 1.7B, first super-resolution 0.8B, final super-resolution 1.8B; paper Table 5). Not stated whether the T5 text encoder or the DeepFloyd IF-II-L first-frame upscaler used by show-1-sr1 are included.

25.2±2.3#89 of 93 video modelsfrom 4 benchmark results across 1 source

$ how it compares

The video models ranked either side of Show-1, out of 139 with a score.

#modelscorewhereRAMreleased
132PikaLab (September 2023) Pika27±2comparison only2023-09
133VideoCrafter1 T2V Tencent AI Lab / HKUST26±2local~8.1GB2023-10
134Show-1 Show Lab / NUS25±2local~83GB2023-10
135LVDM (text-to-video) HKUST / Tencent AI Lab25±3local~6.1GB2023-05
136ModelScope T2V 1.7B Alibaba DAMO Academy25±2local~4.6GB2023-03

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

$ what it's good at

text to video▆▄▃▄▆▅█▃▂▂▂▄ #63 of 68 · 98% of its score · 4 tests

Beats Open-Sora 1.2 here, though they rank higher overall.

testresultvs the others in its rundifficultycountswhy
EvalCrafter 2023 motion quality53.74 / 100 ✎▄····▄██··▄▄ beats 5/7very easy · avg 85.14 / 100▇▇▇▇▇▇▇▇▇▇ 26%at its level
EvalCrafter 2023 visual quality52.19 / 100 ✎▄█·····▄·█▄▄ beats 0/7very easy · avg 87.87 / 100▇▇▇▇▇▇▇▇▇▇ 25%at its level
EvalCrafter 2023 temporal consistency60.83 / 100 ✎▄···▄▄▄▄▄··█ beats 5/7very easy · avg 88.67 / 100▇▇▇▇▇▇▇▇▇▇ 24%at its level
EvalCrafter 2023 text-video alignment62.07 / 100 ✎▂··▂▂·····▂█ beats 7/7very easy · avg 87.86 / 100▇▇▇▇▇▇▇▇▇ 23%at its level

$ where to run it

Other ways to run it

  1. Run it as published · needs ~83.2 GB, NVIDIA only · from github.com/showlab (official)
    git clone https://github.com/showlab/Show-1 && cd Show-1 && pip install -r requirements.txt && python run_inference.py

Official weights and documentation: github.com/showlab ↗

$ how the score was worked out · EvalCrafter: original 2023 te… 98%

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
EvalCrafter: original 2023 text-to-video evaluations4▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 98%CC-BY-4.0
prior · pull toward the middle▇ 2%
weighted average= 25.2

$ 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