whatmodel.sh / video / AnimateDiff SD 1.5 v2 motion adapter

AnimateDiff SD 1.5 v2 motion adapter is an open-weights video model from AnimateDiff authors, released September 2023. Its published size is 0.453B parameters. Licence: apache-2.0.

Catalogue identity notes

Motion adapter requiring a compatible Stable Diffusion 1.5 base; no score borrowed from a complete AnimateDiff pipeline.

Model information sources

· Weights licence ↗ · Parameter count ↗

453M = the AnimateDiff v2 SD 1.5 motion module (mm_sd_v15_v2) only, per the makers' model-zoo table; the Stable Diffusion 1.5 base (UNet, CLIP text encoder, VAE) it plugs into is not included.

No comparable score yet. No reviewed benchmark results yet.

$ how it compares

Scored video models released closest to AnimateDiff SD 1.5 v2 motion adapter, out of 139 with a score.

#modelscorewhereRAMreleased
122LaVie (base T2V) Shanghai AI Lab (Vchitect)32±2local~4.9GB2023-09
127Hotshot-XL Hotshot27±2local~10GB2023-10
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

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

$ where to run it

Other ways to run it

  1. Run it as published · needs ~7.8 GB · from github.com/leejet
    sd-cli -M vid_gen --model models/v1-5-pruned-emaonly.safetensors --motion-module models/mm_sd_v15_v2.ckpt --offload-to-cpu --diffusion-fa -p "a red apple on a wooden table" --cfg-scale 8.0 --sampling-method euler --scheduler discrete -H 384 -W 384 --video-frames 16 --fps 8 --steps 20 -s 42 -o out.avi

Weights and documentation: huggingface.co/guoyww ↗

$ improve this page

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