---
title: mochi-gguf
canonical_url: "https://www.modelscope.cn/models/calcuis/mochi-gguf"
md_url: "https://www.modelscope.cn/models/calcuis/mochi-gguf.md"
repository: calcuis/mochi-gguf
last_updated: 2025-08-10
license: apache-2.0
pipeline_tag: text-to-video-synthesis
tasks:
  - text-to-video-synthesis
base_model:
  - Comfy-Org/mochi_preview_repackaged
base_model_relation: quantized
library_name:
  - gguf
  - pytorch
frameworks:
  - pytorch
language:
  - en
downloads: 998
stars: 0
tags:
  - gguf-node
  - gguf-connector
  - gguf
---

# mochi-gguf

> mochi-gguf - calcuis 在 ModelScope 开源的模型。gguf quantized version of mochi (incl. gguf encoder and gguf vae)🐷🐷🐷 run it straight with gguf-connector opt a gguf file in the current directory to interact with by:

calcuis/mochi-gguf 是 ModelScope 魔搭社区上的text-to-video-synthesis模型，采用 apache-2.0 许可，基于 Comfy-Org/mochi_preview_repackaged 构建。

- **Repository**: calcuis/mochi-gguf
- **License**: apache-2.0
- **Tasks**: text-to-video-synthesis
- **Base model**: Comfy-Org/mochi_preview_repackaged
- **Tags**: gguf-node, gguf-connector, gguf
- **Downloads**: 998
- **Stars**: 0
- **Last updated**: 2025-08-10

Source: https://www.modelscope.cn/models/calcuis/mochi-gguf

---

# **gguf quantized version of mochi (incl. gguf encoder and gguf vae)🐷🐷🐷**
- run it straight with `gguf-connector`
- opt a `gguf` file in the current directory to interact with by:
```
ggc m1
```
>
>GGUF file(s) available. Select which one to use:
>
>1. pig-mochi-q3_k_m.gguf
>2. pig-mochi-q4_1.gguf
>3. pig-mochi-q8_0.gguf
>
>Enter your choice (1 to 3): _
>>>

## **run it with gguf-node via comfyui**
- drag **mochi** to > `./ComfyUI/models/diffusion_models`
- drag **t5xxl** to > `./ComfyUI/models/text_encoders`
- drag **vae** to > `./ComfyUI/models/vae`
- drag demo video (below) to > your browser for workflow

<Gallery />

### **review**
- new tensor fixed version; load faster with **full set gguf** (model + encoder + decoder)
- **upgraded encoder**; from fp16/fp8 to **fp32**; but not affecting the file size and memory consumption; more compatible to old machine
- new **fp32 gguf vae decoder**; similar size to fp16 safetensors; better quality; less ram requirement
- **q2** works but not usable; you could get **q8** [[here](https://huggingface.co/calcuis/pig/blob/main/pig-mochi-q8_0.gguf)] pig architecture

![screenshot](https://raw.githubusercontent.com/calcuis/comfy/master/mochi-gguf.webp)

### **reference**
- base model from [genmo](https://huggingface.co/genmo)
- comfyui from [comfyanonymous](https://github.com/comfyanonymous/ComfyUI)
- pig architecture from [connector](https://huggingface.co/connector)
- gguf-node ([pypi](https://pypi.org/project/gguf-node)|[repo](https://github.com/calcuis/gguf)|[pack](https://github.com/calcuis/gguf/releases))
- gguf-connector ([pypi](https://pypi.org/project/gguf-connector))