---
license: apache-2.0
base_model:
- Wan-AI/Wan2.2-S2V-14B
tags:
- gguf-node
widget:
- text: >-
    a cute anime girl with massive fennec ears and a big fluffy tail wearing a maid outfit
  parameters:
    negative_prompt: blurry ugly bad
  output:
    url: workflow-demo1.webp
- text: >-
    a cute anime girl with massive fennec ears and a big fluffy tail wearing a maid outfit turning around
  parameters:
    negative_prompt: blurry ugly bad
  output:
    url: workflow-demo2.webp
- text: >-
    a conversation between cgg and connector
  parameters:
    negative_prompt: blurry ugly bad
  output:
    url: workflow-conversation.mp4
---
## **gguf quantized version of wan2.2-s2v (all gguf: incl. encoders + vae)**
- drag **wan** to > `./ComfyUI/models/diffusion_models`
- either 1 or 2 below, drag it to > `./ComfyUI/models/text_encoders`
  - option 1: just **umt5xxl** [[3.47GB](https://huggingface.co/calcuis/wan-s2v-gguf/blob/main/umt5xxl_fp32-q4_0.gguf)]
  - option 2: both **cat-umt5xxl** [[3.66GB](https://huggingface.co/calcuis/cat-encoder/blob/main/cat-umt5xxl-q2_k.gguf)] and **tokenizer** [[4.55MB](https://huggingface.co/calcuis/cat-encoder/blob/main/tokenizer-umt5xxl-u8.safetensors)]
- drag **wav2vec2** [[632MB](https://huggingface.co/calcuis/wan-s2v-gguf/blob/main/wav2vec2_large_english_fp32-f16.gguf)] to > `./ComfyUI/models/audio_encoders`
- drag **pig** [[254MB](https://huggingface.co/calcuis/wan-s2v-gguf/blob/main/pig_wan_vae_fp32-f16.gguf)] to > `./ComfyUI/models/vae`

![screenshot](https://raw.githubusercontent.com/calcuis/comfy/master/wan22-s2v.png)

<Gallery />

note: the new `GGUF AudioEncoder Loader` on test; running gguf audio encoder `wav2vec2` w/o ending error msg compare to fp16 safetensors (depends how long of your prompt/video)

## **reference**
- for the lite workflow (save >70% loading time), get the `lite lora` for 4/8-step operation [here](https://huggingface.co/calcuis/wan2-gguf/blob/main/wan2.2_s2v_lite_lora.safetensors)
- or opt to use scaled fp8 e4m3 safetensors `audio encoder` [here](https://huggingface.co/chatpig/encoder/blob/main/wav2vec2_large_english_fp8_e4m3fn.safetensors) and/or fp8 e4m3 `vae` [here](https://huggingface.co/calcuis/wan2-gguf/blob/main/wan_2.1_vae_fp8_e4m3fn.safetensors) (don't even need to switch to native loaders as `GGUF AudioEncoder Loader` and `GGUF VAE Loader` support both gguf and fp8 scaled safetensors files)
- gguf-node ([pypi](https://pypi.org/project/gguf-node)|[repo](https://github.com/calcuis/gguf)|[pack](https://github.com/calcuis/gguf/releases))