Blog | Hugging Face | Playground | Careers
A state of the art video generation model by Genmo.
Mochi 1 preview is an open state-of-the-art video generation model with high-fidelity motion and strong prompt adherence in preliminary evaluation. This model dramatically closes the gap between closed and open video generation systems. We’re releasing the model under a permissive Apache 2.0 license. Try this model for free on our playground.
Install using uv:
git clone https://github.com/genmoai/modelscd models pip install uv uv venv .venvsource .venv/bin/activate uv pip install setuptools uv pip install -e . --no-build-isolation
If you want to install flash attention, you can use:
uv pip install -e .[flash] --no-build-isolation
You will also need to install FFMPEG to turn your outputs into videos.
Use download_weights.py to download the model + decoder to a local directory. Use it like this:
python3 ./scripts/download_weights.py
Or, directly download the weights from Hugging Face or via magnet:?xt=urn:btih:441da1af7a16bcaa4f556964f8028d7113d21cbb&dn=weights&tr=udp://tracker.opentrackr.org:1337/announce to a folder on your computer.
Start the gradio UI with
python3 ./demos/gradio_ui.py --model_dir ""
Or generate videos directly from the CLI with
python3 ./demos/cli.py --model_dir ""
Replace
This repository comes with a simple, composable API, so you can programmatically call the model. You can find a full example here. But, roughly, it looks like this:
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,
)Mochi 1 represents a significant advancement in open-source video generation, featuring a 10 billion parameter diffusion model built on our novel Asymmetric Diffusion Transformer (AsymmDiT) architecture. Trained entirely from scratch, it is the largest video generative model ever openly released. And best of all, it’s a simple, hackable architecture. Additionally, we are releasing an inference harness that includes an efficient context parallel implementation.
Alongside Mochi, we are open-sourcing our video AsymmVAE. We use an asymmetric encoder-decoder structure to build an efficient high quality compression model. Our AsymmVAE causally compresses videos to a 128x smaller size, with an 8x8 spatial and a 6x temporal compression to a 12-channel latent space.
| Params Count | Enc Base Channels | Dec Base Channels | Latent Dim | Spatial Compression | Temporal Compression |
|---|---|---|---|---|---|
| 362M | 64 | 128 | 12 | 8x8 | 6x |
An AsymmDiT efficiently processes user prompts alongside compressed video tokens by streamlining text processing and focusing neural network capacity on visual reasoning. AsymmDiT jointly attends to text and visual tokens with multi-modal self-attention and learns separate MLP layers for each modality, similar to Stable Diffusion 3. However, our visual stream has nearly 4 times as many parameters as the text stream via a larger hidden dimension. To unify the modalities in self-attention, we use non-square QKV and output projection layers. This asymmetric design reduces inference memory requirements. Many modern diffusion models use multiple pretrained language models to represent user prompts. In contrast, Mochi 1 simply encodes prompts with a single T5-XXL language model.
| Params Count | Num Layers | Num Heads | Visual Dim | Text Dim | Visual Tokens | Text Tokens |
|---|---|---|---|---|---|---|
| 10B | 48 | 24 | 3072 | 1536 | 44520 | 256 |
The model requires at least 4 H100 GPUs to run. We welcome contributions from the community to reduce this requirement.
Genmo video models are general text-to-video diffusion models that inherently reflect the biases and preconceptions found in their training data. While steps have been taken to limit NSFW content, organizations should implement additional safety protocols and careful consideration before deploying these model weights in any commercial services or products.
Under the research preview, Mochi 1 is a living and evolving checkpoint. There are a few known limitations. The initial release generates videos at 480p today. In some edge cases with extreme motion, minor warping and distortions can also occur. Mochi 1 is also optimized for photorealistic styles so does not perform well with animated content. We also anticipate that the community will fine-tune the model to suit various aesthetic preferences.
ComfyUI-MochiWrapper adds ComfyUI support for Mochi. The integration of Pytorch's SDPA attention was taken from their repository.
@misc{genmo2024mochi,
title={Mochi},
author={Genmo Team},
year={2024}
} 
