Downloads · 30 days
12
17% of all-time downloads
TheMindExpansionNetwork/swing-u-sinnerz
swing-u-sinnerz is a text-to-video model from TheMindExpansionNetwork. Use it when you need video from a text prompt. It is set up for diffusers. The card lists the license as other.
This is a fine-tuned version of ltx-2-19b-dev.safetensors trained on custom data.
Downloads · 30 days
12
17% of all-time downloads
All-time downloads
72
Public
Repo size
468 MB
Likes
0
Public
Click a slice to open those files.
.safetensors428 MB · 91%
From the Hugging Face model README
This is a fine-tuned version of ltx-2-19b-dev.safetensors trained on custom data.
ltx-2-19b-dev.safetensors <br><details style="max-width: 300px; margin: auto;"><summary>Prompt</summary>A black-and-white 1930s rubber-hose cartoon character with stretchy limbs walks down a crooked city street at night under a flickering streetlamp, pausing to glance at a shadow that slowly crawls along a brick wall behind them. The characters eyes widen and their mouth stretches in a startled expression, then they shuffle backward with bouncy steps as the shadow grows larger. The audio includes a soft projector hum and film crackle, light tap-tap footsteps on pavement, a quiet spooky jazz riff from muted brass and upright bass, and a short cartoon whoosh as the shadow slides across the wall.</details> | <br><details style="max-width: 300px; margin: auto;"><summary>Prompt</summary>A black-and-white 1930s rubber-hose cartoon character sits at a small upright piano in a surreal underworld room with warped picture frames and swaying candles, playing a simple jazzy rhythm while their head and shoulders bob in time. As the melody continues, the room’s shadows stretch and bend in odd shapes and the character looks left and right with an uneasy grin, then pulls their hands back as if something is moving inside the piano. The audio features soft piano notes with a light swing feel, subtle projector hiss and film crackle, gentle creaks from the room, and brief cartoon squeaks timed to the character’s movements.</details> |
This model is designed to be used with the LTX-2 (Lightricks Audio-Video) pipeline.
In order to use the trained LoRA in ComfyUI, follow these steps:
.safetensors file) to the models/loras folder in your ComfyUI installation.You can find reference Text-to-Video (T2V) and Image-to-Video (I2V) workflows in the official LTX-2 repository.
Example prompts used during validation:
A black-and-white 1930s rubber-hose cartoon character with stretchy limbs walks down a crooked city street at night under a flickering streetlamp, pausing to glance at a shadow that slowly crawls along a brick wall behind them. The characters eyes widen and their mouth stretches in a startled expression, then they shuffle backward with bouncy steps as the shadow grows larger. The audio includes a soft projector hum and film crackle, light tap-tap footsteps on pavement, a quiet spooky jazz riff from muted brass and upright bass, and a short cartoon whoosh as the shadow slides across the wall.A black-and-white 1930s rubber-hose cartoon character sits at a small upright piano in a surreal underworld room with warped picture frames and swaying candles, playing a simple jazzy rhythm while their head and shoulders bob in time. As the melody continues, the room’s shadows stretch and bend in odd shapes and the character looks left and right with an uneasy grin, then pulls their hands back as if something is moving inside the piano. The audio features soft piano notes with a light swing feel, subtle projector hiss and film crackle, gentle creaks from the room, and brief cartoon squeaks timed to the character’s movements.This model inherits the license of the base model (ltx-2-19b-dev.safetensors).