Boba's MMAudio Workflow — Comfy Core
Today I'm providing an easy to use MMAudio workflow.
Runtime profile
- Access
- External workflow
- Status
- Listed
- Version
- Boba's MMAudio ComfyCore v1.json
- Source
- Civitai
- Updated
- Oct 13, 2026
Source description
Today I'm providing an easy to use MMAudio workflow.
This workflow allows you to add audio to your videos, an image, or to simply generate audio via text prompting through MMAudio.
Notes are included with this workflow explaining some developer notes on MMAudio, where to download the required files, alongside a quality of life option to add additional frames to your video if needed, prompting tips, and an option to merge multiple MMAudio clips together.
Currently I'm providing two versions.
Comfy Core - This workflow uses the minimum amount of custom nodes required to run MMAudio.
MMAudio Plus - This workflow uses a few different custom nodes to add more convenience.
Note: The example videos utilize both MMAudio for sound effects and Seed Voice Conversion for voices.
Seed Voice Conversion's github can be found here and there's custom nodes available through ComfyUI's manager. https://github.com/billwuhao/ComfyUI_Seed-VC
What you need
- Model files
- 3.7 GiB known subtotal
Remaining hardware or download-size estimates are not available yet.
Requirements
Models
4Text encoder · 3.68 GB · Hugging Face · KrazyEyezKillah/comfyui-nsfw-model-pack
mmaudio_large_44k_v2_fp32.safetensors
PossibleCheckpoint · Unknown
mmaudio_synchformer_fp32.safetensors
PossibleCheckpoint · Unknown
mmaudio_vae_44k_fp32.safetensors
PossibleVAE · Unknown
Node packs
1- Found
Node pack · Registry
Sources
1 source
Sources
1 sourceSource excerpts
1 excerpt
Source excerpts
1 excerptSource context: 771 downloads · Type Workflows · Base model Other
