void-model-mlx / README.md

dgrauet

Refresh the card so its own manifest size line is current

0502025 verified 16 days ago

preview code

Raw

History Blame Contribute Delete

1.45 kB

metadata

library_name: mlx
license: apache-2.0
base_model: netflix/void-model
tags:
  - mlx
  - mlx-forge
  - apple-silicon
  - safetensors

dgrauet/void-model-mlx

MLX format conversion of netflix/void-model.

Converted with mlx-forge.

Usage

These weights can be used with void-model-mlx:

git clone https://github.com/dgrauet/VideoX-Fun-mlx.git
export VIDEOX_FUN_MLX_PATH=/path/to/VideoX-Fun-mlx
pip install mlx opencv-python-headless pillow numpy sentencepiece

python -m void_mlx.infer \
    --sample sample/BigBen \
    --pass1 weights/void_pass1.safetensors \
    --pass2 weights/void_pass2.safetensors \
    --base-model /path/to/CogVideoX-Fun-V1.5-5b-InP-mlx-q8 \
    --steps 30 --max-frames 13 --height 352 --width 624 \
    --low-ram --output result.gif

Related Projects

void-model-mlx (inference): https://github.com/dgrauet/void-model-mlx
VideoX-Fun-mlx (engine): https://github.com/dgrauet/VideoX-Fun-mlx
Base model weights: https://huggingface.co/dgrauet/CogVideoX-Fun-V1.5-5b-InP-mlx
q8 variant: https://huggingface.co/dgrauet/void-model-mlx-q8
q4 variant: https://huggingface.co/dgrauet/void-model-mlx-q4

Files

config.json (376.00 B)
split_model.json (1.18 KB)
void_pass1.safetensors (10.38 GB)
void_pass2.safetensors (10.38 GB)