0:00
/0:02

In this guide you'll get started with one of the latest and greatest AI video models, Mochi, within ComfyUI.

Mochi excels turning text into video (text2video) with high quality motion and prompt adherence.

Did we mention that ComfyUI is open source? All you need is a local install of ComfyUI or a cloud solution like ThinkDiffusion.

The workflow can be downloaded below, all you need to do is drag & drop it into your ComfyUI and follow the steps below.

How to run Mochi text-to-video in ComfyUI

Installation guide

ThinkDiffusion-StableDiffusion-ComfyUI-Mochi
DOWNLOAD THE FILE HERE
ThinkDiffusion-StableDiffusion-ComfyUI-Mochi.json
15 KB
download-circle
💡
Download the workflow and drag & drop it into your ComfyUI window, whether locally or on ThinkDiffusion. If you're using ThinkDiffusion, it's necessary to use at minimum the Turbo 24gb machine, but we do recommend the Ultra 48gb machine.

Custom Nodes

If there are red nodes in the workflow, it means that the workflow lacks the certain required nodes. Install the custom nodes in order for the workflow to work.

  1. Go to ComfyUI Manager  > Click Install Missing Custom Nodes
ThinkDiffusion StableDiffusion ComfyUI Mochi install the missing custom node
  1. Check the list below if there's a list of custom nodes that needs to be installed and click the install.
ThinkDiffusion StableDiffusion ComfyUI Mochi check the list of missing custom nodes

Models

Download the recommended models (see list below) using the ComfyUI manager and go to Install models. Refresh or restart the machine after the files have downloaded.

  1. Go to ComfyUI Manager  > Click Model Manager
ThinkDiffusion StableDiffusion ComfyUI Mochi install models using the model manager
  1. When you find the exact model that you're looking for, click install and make sure to press refresh when you are finished.
ThinkDiffusion StableDiffusion ComfyUI Mochi check the list of missing models

Model Path Source

Use the model path source if you prefer to install the models using model's link address and paste into ThinkDiffusion MyFiles using upload URL.

Model Name Model Link Address
mochi_preview_dit_bf16.safetensors
📋 Copy Path
t5xxl_fp16.safetensors
📋 Copy Path
mochi_vae.safetensors
📋 Copy Path

Guide Table for Upload

Recommended Models

Node’s Value Name

Node

ThinkDiffusion Upload File Directory

mochi_preview_dit_bf16.safetensors

unet_name

Load Diffusion Model

…comfyui/models/unet/

t5xxl_fp16.safetensors

clip_name

Load CLIP

…comfyui/models/clip/

mochi_vae.safetensors

vae_name

Load VAE

…comfyui/models/vae/

Step-by-step guide for Mochi in ComfyUI

STEPS Default Nodes to Follow
1. Set the Models ThinkDiffusion StableDiffusion ComfyUI Mochi set the models
2. Write a Prompt ThinkDiffusion StableDiffusion ComfyUI Mochi write a prompt
3. Set the Video Size

Set the resolution to 480 as this is the only compatible size for the Mochi 1 models.		 ThinkDiffusion StableDiffusion ComfyUI Mochi set the video size
4. Set the Settings

Set the steps between 50 to 100. I found the sweet spot around 75 steps		 ThinkDiffusion StableDiffusion ComfyUI Mochi set the settings
5. Check the Video Output ThinkDiffusion StableDiffusion ComfyUI Mochi check the video output

Notes

💡
Mochi 1 is excels with simple generations like landscapes, animals, nature and other non-human based subject. Mochi 1 is also optimized for photorealistic styles so it doesn't perform as well with animated content. It has some minor distortions with if the prompt contains a extreme motion.
💡
Mochi 1 models were trained using 480 resolution. If you set your video rendering beyond that resolution, the generation may stop.

Mochi text-to-video in ComfyUI Examples

0:00
/0:02
0:00
/0:02
0:00
/0:02
0:00
/0:02

CALM FOREST

  • Prompt - A calming nature walk through a beautiful autumn forest.
  • Seed - 926345488871431
  • Steps 75, CFG 4.5
  • Euler , Simple
  • Mochi_preview_dit_bf16.safetensors
  • t5xxl_fp16.safetensors

EMERGING ANTS

  • Prompt - The underground world of ants building their complex colony.
  • Seed - 654412196420470
  • Steps 75, CFG 4.5
  • Euler , Simple
  • Mochi_preview_dit_bf16.safetensors
  • t5xxl_fp16.safetensors

WATERWORLD

  • Prompt - The vibrant nightlife of a coral reef under a full moon.
  • Seed - 568746619838427
  • Steps 75, CFG 4.5
  • Euler , Simple
  • Mochi_preview_dit_bf16.safetensors
  • t5xxl_fp16.safetensors

TREE TIMELAPSE

  • Prompt - A time-lapse of a seed growing into a towering tree in a forest.
  • Seed - 980178990217739
  • Steps 75, CFG 4.5
  • Euler , Simple
  • Mochi_preview_dit_bf16.safetensors
  • t5xxl_fp16.safetensors

If you’re having issues with installation or slow hardware, you can try any of these workflows on a more powerful GPU in your browser with ThinkDiffusion.

TRY THINKDIFFUSION

If you enjoy ComfyUI and you want to test out creating awesome animations, then feel free to check out this AnimateDiff tutorial here. And have fun out there with your noodles!