Skip to content
WorkflowListedExternal
Download

Long Video Subgraphs Wan 2.2 — Long Wan 2.2 MMAudio

mmaudio added

Sourcecivitai.com/models/2038030
TaskMultiCapabilitiesImage to image · UpscalingFamilyWanVersionRelease 2308409 — Long Wan 2.2 MMAudio.jsonEstimated VRAMNot yet characterizedUnratedgrade

Runtime profile

Access
External workflow
Status
Listed
Version
Long Wan 2.2 MMAudio.json
Source
Civitai
Updated
Nov 3, 2026

What you need

Model files
10 GiB known subtotal

Remaining hardware or download-size estimates are not available yet.

Requirements

Models

11

Node packs

5

  • comfyui_controlnet_aux

    Possible

    Node pack · Registry

  • comfyui-kjnodes

    Found

    Node pack · GitHub · kijai/ComfyUI-KJNodes

  • Found

    Node pack · Registry

  • comfyui-videohelpersuite

    Found

    Node pack · GitHub · Kosinkadink/ComfyUI-VideoHelperSuite

  • comfyui-wanvideowrapper

    Found

    Node pack · Registry

AI-generated commentary

AI-generated explanation based on source and configuration details. Suggestions are clearly labeled.

Purpose

  • Long Video Subgraphs Wan 2.2 is for generating longer Wan 2.2 videos in 81-frame batches. It uses a prompt and seed at each subgraph and produces longer videos.

Flow

  • The workflow is organized around subgraphs and 81-frame batches for longer-video generation.

Inputs and outputs

  • Inputs are a prompt and seed for each subgraph; the output is a longer Wan 2.2 video generated in 81-frame batches.

Setup

  • The listed base model is Wan Video 2.2 I2V-A14B.

  • The listed node-pack versions are comfyui-kjnodes 1.1.6, comfyui-videohelpersuite 1.7.2, and comfyui-wanvideowrapper 1.3.5.

Dependencies

  • The listed node packs are comfyui_controlnet_aux, comfyui-kjnodes, comfyui-mmaudio, comfyui-videohelpersuite, and comfyui-wanvideowrapper.

  • Listed model files include umt5_xxl_fp8_e4m3fn_scaled.safetensors, lightx2v_I2V_14B_480p_cfg_step_distill_rank64_bf16.safetensors, wan2.2_i2v_low_noise_14B_fp8_scaled.safetensors, and wan2.2_i2v_high_noise_14B_fp8_scaled.safetensors.

  • Additional listed files are mmaudio_vae_44k_fp16.safetensors, mmaudio_large_44k_v2_fp16.safetensors, mmaudio_synchformer_fp16.safetensors, wan_2.1_vae.safetensors, and apple_DFN5B-CLIP-ViT-H-14-384_fp16.safetensors.

Configuration

  • Set a prompt and seed at each subgraph.

    Suggestion · not verified

Cautions

  • Before use, verify that these listed files are available: mmaudio_vae_44k_fp16.safetensors, mmaudio_large_44k_v2_fp16.safetensors, mmaudio_synchformer_fp16.safetensors, wan_2.1_vae.safetensors, and apple_DFN5B-CLIP-ViT-H-14-384_fp16.safetensors.

    Suggestion · not verified

  • Verify which package provides comfyui_controlnet_aux before use; no provider is identified for it.

    Suggestion · not verified

Does this need to be edited?

Sign in to send an edit request.

Sources

1 source
CivitaiPrimaryVersion 2308409
civitai.com/models/2038030

Verified Nov 3

Workflow instructions

Author notes from Long Wan 2.2 MMAudio.json · selected release

Wan 2.2 i2V models, high and low with Lightning loras. Use what works best for you though.

Author notes from Long Wan 2.2 MMAudio.json · selected release

typical vae and text encorder

Download files

Model artifacts

2 artifacts

longVideoSubgraphsWan_longWan22Mmaudio.zip

zip · 19.8 KB · SHA-256 b881b8bc3c51…edf0 · Civitai

Download file

Long Wan 2.2 MMAudio.json

json · 206 KB · SHA-256 fae5a55042f2…cc7d

Download file

Source excerpts

3 excerpts
NoteNov 3civitai.com/models/2038030

Source context: 1270 downloads · Type Workflows · Base model Wan Video 2.2 I2V-A14B

NoteOct 2civitai.com/models/2038030

Source context: 1237 downloads · Type Workflows · Base model Wan Video 2.2 I2V-A14B

NoteJul 19civitai.com/models/2038030

Source context: 1162 downloads · Type Workflows · Base model Wan Video 2.2 I2V-A14B