Modelpublic

FastVideo/LingBot-Video-MoE-30B-A3B-Diffusers

sourceHugging Faceapache-2.0updated 2mo agoView on Hugging Face

1likes

Model Card

LingBot-Video MoE 30B-A3B for FastVideo

This repository contains the released LingBot-Video MoE 30B-A3B base model and its second-stage refiner in the component layout loaded natively by FastVideo.

Usage

Use a FastVideo revision containing LingBot-Video support:

python

from fastvideo import VideoGenerator

generator = VideoGenerator.from_pretrained(
    "FastVideo/LingBot-Video-MoE-30B-A3B-Diffusers",
    num_gpus=8,
    sp_size=8,
    use_fsdp_inference=True,
    refine_enabled=True,
    text_encoder_cpu_offload=True,
)
result = generator.generate({
    "prompt": "A red fox runs through fresh snow at sunrise.",
    "output": {"output_path": "outputs/lingbot-video", "save_video": True},
})
generator.shutdown()

The released preset generates a 121-frame 832×480 base video and optionally refines it to 1920×1088. Set refine_enabled=False when loading the model to run the base MoE transformer without the refiner.

Provenance

—Original model: https://huggingface.co/robbyant/lingbot-video-moe-30b-a3b
—Original code: https://github.com/robbyant/lingbot-video
—FastVideo port: https://github.com/hao-ai-lab/FastVideo/pull/1595

The original model and this converted package are distributed under the Apache-2.0 license declared by the released checkpoint.