<p>I ENCOURAGE YOU TO MAKE AND POST SOME VIDEOS, TRAINING THIS WAS FAR FROM FREE!</p><p>AB TESTS 4 and 8 steps:</p><div data-youtube-video><iframe width="640" height="480" allowfullscreen="true" autoplay="false" disablekbcontrols="false" enableiframeapi="false" endtime="0" ivloadpolicy="0" loop="false" modestbranding="false" origin playlist src="https://www.youtube.com/embed/NliyTVQi_m8?rel=1" start="0"></iframe></div><p></p><div data-youtube-video><iframe width="640" height="480" allowfullscreen="true" autoplay="false" disablekbcontrols="false" enableiframeapi="false" endtime="0" ivloadpolicy="0" loop="false" modestbranding="false" origin playlist src="https://www.youtube.com/embed/ZHHu_WrkuL8" start="0"></iframe></div><h2 id="short-model-description">Short model description</h2><p><code>frieren_000003000.safetensors</code> is a rank-32 motion and native-audio LoRA for</p><p>the LTX 2 family, tested with the LTX 2.3 22B distilled image-to-video workflow.</p><p>It is intended to animate an already-correct first frame into restrained anime</p><p>character performance: breathing, blinking, gaze changes, conversational</p><p>reactions, walking, sequential dialogue, and quiet environmental movement.</p><p>It works best when the prompt explicitly identifies every visible character, assigns every</p><p>character a speech state, and uses <code>Then</code>, <code>responds</code>, and <code>replies</code> to express</p><p>speaker order.</p><p>This is <strong>i2v LoRA</strong>. It does not create the initial</p><p>Frieren/Fern/Stark image and cannot repair a wrong first-frame identity. Generate</p><p>or supply the correct guide image first, then use this LoRA for LTX motion and</p><p>native audio.</p><h2 id="recommended-settings">Recommended settings</h2><p>Ltx 2.3 dev distilled</p><p>8 steps</p><p>i2v</p><p>1080p (1920*1080)</p><h2 id="what-it-is-good-at">What it is good at</h2><p>- Calm anime I2V performance from a strong first-frame guide.</p><p>- Frieren, Fern, and Stark reacting naturally in the same shot.</p><p>- One visible speaker while the other characters listen silently.</p><p>- One, two or three speakers talking in an explicitly ordered sequence. Also by prompting "Fern is speaking off-screen..." you can often make it happen.</p><p>- Native LTX ambience and speech generated with the video.</p><p>- Subtle breathing, blinking, gaze changes, hair/cloth movement, and wind.</p>

Files with SHA256: 5c4e2af5cc680846b59b91b8ca001f2fe4f38f3f9e72960332a9be9536f6b114