genmo/mochi-1-preview

Instalación desde el repositorio de Genmo:

git clone https://github.com/genmoai/models
cd models
pip install uv
uv venv .venv
source .venv/bin/activate
uv pip install setuptools
uv pip install -e . --no-build-isolation

Para instalar Flash Attention:

uv pip install -e .[flash] --no-build-isolation

Ejecutar la interfaz Gradio:

python3 ./demos/gradio_ui.py --model_dir ""

Generar videos desde CLI:

python3 ./demos/cli.py --model_dir ""

Uso programático con la API de Genmo:

from genmo.mochi_preview.pipelines import (
    DecoderModelFactory,
    DitModelFactory,
    MochiSingleGPUPipeline,
    T5ModelFactory,
    linear_quadratic_schedule,
)

pipeline = MochiSingleGPUPipeline(
    text_encoder_factory=T5ModelFactory(),
    dit_factory=DitModelFactory(
        model_path=f"{MOCHI_DIR}/dit.safetensors",
        model_dtype="bf16"),
    decoder_factory=DecoderModelFactory(
        model_path=f"{MOCHI_DIR}/vae.safetensors",
    ),
    cpu_offload=True,
    decode_type="tiled_full",
)

video = pipeline(
    height=480,
    width=848,
    num_frames=31,
    num_inference_steps=64,
    sigma_schedule=linear_quadratic_schedule(64, 0.025),
    cfg_schedule=[4.5] * 64,
    batch_cfg=False,
    prompt="your favorite prompt here ...",
    negative_prompt="",
    seed=12345,
)

Uso con Diffusers:

pip install git+https://github.com/huggingface/diffusers.git

import torch
from diffusers import MochiPipeline
from diffusers.utils import export_to_video

pipe = MochiPipeline.from_pretrained("genmo/mochi-1-preview")
pipe.enable_model_cpu_offload()
pipe.enable_vae_tiling()

prompt = "Close-up of a chameleon's eye, with its scaly skin changing color. Ultra high resolution 4k."

with torch.autocast("cuda", torch.bfloat16, cache_enabled=False):
    frames = pipe(prompt, num_frames=84).frames[0]

export_to_video(frames, "mochi.mp4", fps=30)

Variante bf16 para ahorrar memoria:

import torch
from diffusers import MochiPipeline
from diffusers.utils import export_to_video

pipe = MochiPipeline.from_pretrained(
    "genmo/mochi-1-preview",
    variant="bf16",
    torch_dtype=torch.bfloat16
)
pipe.enable_model_cpu_offload()
pipe.enable_vae_tiling()

prompt = "Close-up of a chameleon's eye, with its scaly skin changing color. Ultra high resolution 4k."
frames = pipe(prompt, num_frames=84).frames[0]
export_to_video(frames, "mochi.mp4", fps=30)