Workflow and models for "very simple" movements?

via

StableDiffusion

Every time i try to create simple movements via LTX or Wan, the output is not even close to what i want to achieve. Like i prompt around a simple movement, like a girl literally only looking into a camera, but in the output the girl always rapidly open and closes her eyes, opens her mouth, starts talking and does other weird ass movements. What is the best way to create some simple natural movements with 16GB Vram?

More from StableDiffusion

Is this result possible using local models? What could they be using?

StableDiffusion·4

Phosphene — local video and audio generation for Apple Silicon ( LTX2.3 )

StableDiffusion·3

Comfyui Video Combine Plus

StableDiffusion·4

He created another self : Sci-Fi Short Film

StableDiffusion·4

Anima seems to do impressively well on json formatted prompt

StableDiffusion·2

Amazing style made using Ernie Image Turbo

StableDiffusion·4