Vimoo-1.0

Vimoo-1.0

Makes video from a text prompt, or from a reference image plus text, with a separate 11.7B model for each job. Suits creators and marketing teams who want short clips.

Made by Vimoo-AI on Hugging Face. These notes are Mindket's, written from its model page and model card; the model's own files and license are on Hugging Face.

About

What it is: Vimoo-1.0 makes video from a text prompt, or from a reference image plus text, using a separate model of about 11.7B parameters for each job. It is the best-quality member of the Vimoo 1.0 family. The same family is also offered as an online API, with ComfyUI and Apple Silicon MLX versions.

Who it suits: Video creators, marketers and designers who want short clips from a written description or a still image.

Industry | Who uses it | How they use it | Example
Marketing | Brand teams | Make product showcase videos from product pictures | A vertical promo clip built from a handful of product images
Film and media | Storytellers | Turn scene descriptions into short clips | A 5-second 720P clip of a scene with a camera move
Design | Motion designers | Animate a reference image with text direction | A still image brought to life with a mood and a camera move

If you have no GPU with about 23.4 GB of room, you can use the online API instead.

Where to use it:

• Diffusers in Python
• ComfyUI, using the workflows from the ComfyUI-Vimoo repo
• Apple Silicon Macs, through the MLX version
• The Vimoo online API, with no local setup

Use case | Who | What they get back
Text to video | Creators | A 5-second 720P clip from your description
Image to video | Designers | A clip that keeps the look and subject of your reference image
E-commerce video | Online sellers | A 20-second 9:16 720P product video from 1 to 9 images

How to set it up:

• Python with pip
• A GPU with room for about 23.4 GB of BF16 weights
• For ComfyUI or Apple Silicon, the separate repos and weights the card points to

1. Install the Hugging Face hub package.
2. Download the model files.
3. Follow the Quick Start section of the model card to build the Diffusers pipeline.

pip install -U huggingface_hub
hf download Vimoo-AI/Vimoo-1.0

How to use it: Write a prompt that describes the scene, the camera moves and the mood. For image to video, add a reference image. The card recommends running prompts through Vimoo-PE, or following its Prompt Guide, because prompt quality shapes the result.

Tool | What it does
Vimoo-PE | Refines your instructions into a form the generator handles well
Vimoo-Base | Generates the 720p video
Vimoo-Enhancer | Regenerates the 720p video at 1080p

Example: You describe a scene with a slow camera move and a calm mood, and ask for a 5-second clip. A 720P video of that scene comes back.

For newcomers and experts: If you are new, try the online API or the ComfyUI workflow first, since both skip the Python setup. Start with one short prompt and a single scene.

The full system has three stages. A prompt module structures the request, a base model makes 720p video, and an optional enhancer regenerates it at 1080p with repair of defects in detail and motion. The trade-off is extra compute for the enhancer, which the API treats as optional.

Your prompt or reference image
Vimoo-PE structures the request
Vimoo-Base makes 720p video
Vimoo-Enhancer can lift it to 1080p

Good to know:

• Each task has its own model, one for text to video and one for image to video
• The e-commerce variant takes 1 to 9 images, with optional titles and selling points
• License: Apache 2.0
• Cost to run: local runs need a large GPU, and the online API is a separate Vimoo service whose price the card does not name
• Data it reaches: your prompts and images, which go to Vimoo if you use the API
• Not for: long clips, since the listed variants make 5 seconds, or 20 seconds for e-commerce

Title: Clip length by variant
Vimoo-1.0: 5 seconds
Vimoo-1.0-Ecom: 20 seconds
Vimoo-1.0-Lite: 5 seconds

On Hugging Face, 13 November 2026

• Task: text to video
• Library: diffusers
• Parameters: 11.7B
• Weights: about 23.4 GB in BF16; about 5.8 GB at 4-bit (estimate)
• Formats: Safetensors
• Downloads in the last 30 days: 57
• Likes: 18
• License: apache-2.0
• Quantized versions: 1
• Tags: diffusers, safetensors, vimoo, image-to-video, text-to-video

Links

• Model: https://huggingface.co/Vimoo-AI/Vimoo-1.0
• Files: https://huggingface.co/Vimoo-AI/Vimoo-1.0/tree/main

Get the model

Download its files with Hugging Face's command line:

pip install -U huggingface_hub
hf download Vimoo-AI/Vimoo-1.0

Then follow the usage in its model card. Nothing in this kit runs by itself.