<h2 id="model-description-7ud5d1tfl">Model Description</h2>I'm glad to share with you my latest experiment, a basic camera motion LoRA trained with 32-frames on an Animatediff v2 model. The training process utilized Kijai's <a target="_blank" rel="ugc" href="https://github.com/kijai/ComfyUI-ADMotionDirector">AnimateDiff Motion Director ComfyUI nodes</a>. This LoRA has been tested exclusively in ComfyUI with Kosinkadink's <a target="_blank" rel="ugc" href="https://github.com/Kosinkadink/ComfyUI-AnimateDiff-Evolved">AnimateDiff-Evolved nodes</a>, however it may be compatible with other AnimateDiff implementations.<h2 id="how-to-use-77bdpd9kv">How to use</h2><ul><li>copy the motion lora files into your "ComfyUI\models\animatediff_motion_lora" folder</li><li>Use the "Load Animatediff LoRA" node</li><li>set the context_length parameter to 32 and context_overlap to 8 (it might work with other values but this is how I tested it)</li></ul><h2 id="settings-and-recommendations-9zutgqi2e">Settings and recommendations</h2><h3 id="v2-w2cbuiqul">V2</h3><ul><li>example workflow: <a target="_blank" rel="ugc" href="https://civarchive.com/articles/7510">https://civarchive.com/articles/7510</a></li></ul><ul><li>Use it with strength: 0.9</li><li>This motion LoRA is finetuned on a v2 motion model. It works also with AnimateLCM but don't work with v3 models</li><li>It works with SD 1.5 checkpoints</li><li>Animatediff v3 adapter LoRA is recommended regardless it is a v2 model.</li><li>If you want more motion try increasing the scale multival or lower the strength of the v3 adapter LoRA</li><li>Try playing with the strength of<ul><li>motion LoRA</li><li>scale multival</li><li>v3_adapter lora</li></ul>like increasing the scale multival and lowering the lora strength.</li><li>I you can use IP-Adapter, however it reduces the effect of the mLORAs, so play with the strength. "Ease in out" weight type is a good starting point.</li></ul><h3 id="v1-0x69tk64d">V1</h3><ul><li>example workflow: <a target="_blank" rel="ugc" href="https://civarchive.com/articles/6626">https://civarchive.com/articles/6626</a></li></ul><ul><li>Use it with strength: 0.8</li><li>This motion LoRA is finetuned on a v2 motion model. It works also with AnimateLCM but don't work with v3 models</li><li>It works with SD 1.5 checkpoints</li><li>Animatediff v3 adapter LoRA is recommended regardless it is a v2 model.</li><li>If you want more motion try increasing the scale multival or lower the strength of the v3 adapter LoRA</li><li>Try playing with the strength of<ul><li>motion LoRA</li><li>scale multival</li><li>v3_adapter lora</li></ul>like increasing the scale multival and lowering the lora strength.</li><li>I recommend using IP-Adapter, however it reduces the effect of the mLORAs, so play with the strength. "Ease in out" weight type is a good starting point.</li></ul>

Files with SHA256: 072b08b4889f0ddecd17797462b6338f8b18cde2b485842aa32343bea4e84a03