Home Creators Posts Import Register Favorites Logout

Content

Video : https://youtu.be/F9LFAeUbBIs

Discover the future of AI-generated music with HeartMuLa, the groundbreaking open-source music model that just dropped in January 2026—and now runs seamlessly inside ComfyUI! In this video, we walk you through installing and using custom nodes that unlock HeartMuLa’s full potential: generate full-length, high-fidelity songs from simple lyrics and style tags, or transcribe any audio—spoken or sung—with impressive accuracy. Whether you’re creating background tracks for videos, experimenting with AI vocals, or syncing lyrics to animated characters, this workflow puts powerful, offline, and free music generation right at your fingertips.

This tutorial is perfect for AI creators, indie filmmakers, content producers, musicians, and ComfyUI power users who want to integrate dynamic, customizable audio into their visual workflows. You don’t need expensive subscriptions or cloud services—just a local setup and a bit of curiosity. If you’ve used tools like Suno, Stable Audio, or LTX for video, you’ll love how HeartMuLa adds expressive, controllable music and transcription capabilities directly into your node-based pipeline, making it easier than ever to build synchronized audiovisual experiences.

Why does this matter? Because HeartMuLa represents a major leap in open-source music AI—offering multi-language support, section-level control, and studio-quality output without relying on corporate APIs. With its 3-billion-parameter LLM backbone and efficient HeartCodec architecture, it delivers both creativity and technical precision. As AI-generated media evolves, having tools like this in your local toolkit ensures creative freedom, faster iteration, and total ownership over your content. This isn’t just another music generator—it’s a new playground for the next generation of digital storytellers.

Resources:

HeartMuLa official GITHUB https://github.com/HeartMuLa/heartlib

HeartMuLa_ComfyUI https://github.com/benjiyaya/HeartMuLa_ComfyUI

Models:

Go to ComfyUI/models

Use HuggingFace Cli download model weights.

type :

hf download HeartMuLa/HeartMuLaGen --local-dir ./HeartMuLa

hf download HeartMuLa/HeartMuLa-oss-3B --local-dir ./HeartMuLa/HeartMuLa-oss-3B

hf download HeartMuLa/HeartCodec-oss --local-dir ./HeartMuLa/HeartCodec-oss

hf download HeartMuLa/HeartTranscriptor-oss --local-dir ./HeartMuLa/HeartTranscriptor-oss

Files

HeartMula Music AI Model In ComfyUI - The Next Level AI Music And It's Fun!

Discover the future of AI-generated music with HeartMuLa, the groundbreaking open-source music model that just dropped in January 2026—and now runs seamlessly inside ComfyUI! In this video, we walk you through installing and using custom nodes that unlock HeartMuLa’s full potential: generate full-length, high-fidelity songs from simple lyrics and style tags, or transcribe any audio—spoken or sung—with impressive accuracy. Whether you’re creating background tracks for videos, experimenting with AI vocals, or syncing lyrics to animated characters, this workflow puts powerful, offline, and free music generation right at your fingertips. This tutorial is perfect for AI creators, indie filmmakers, content producers, musicians, and ComfyUI power users who want to integrate dynamic, customizable audio into their visual workflows. You don’t need expensive subscriptions or cloud services—just a local setup and a bit of curiosity. If you’ve used tools like Suno, Stable Audio, or LTX for video, you’ll love how HeartMuLa adds expressive, controllable music and transcription capabilities directly into your node-based pipeline, making it easier than ever to build synchronized audiovisual experiences. Why does this matter? Because HeartMuLa represents a major leap in open-source music AI—offering multi-language support, section-level control, and studio-quality output without relying on corporate APIs. With its 3-billion-parameter LLM backbone and efficient HeartCodec architecture, it delivers both creativity and technical precision. As AI-generated media evolves, having tools like this in your local toolkit ensures creative freedom, faster iteration, and total ownership over your content. This isn’t just another music generator—it’s a new playground for the next generation of digital storytellers. Resources: HeartMuLa official GITHUB https://github.com/HeartMuLa/heartlib HeartMuLa_ComfyUI https://github.com/benjiyaya/HeartMuLa_ComfyUI Thank you HeartMuLa Team created a great and fun AI :) -------------------------------------------------------------------------------------------------------------------------------- Local Workstation GPU : https://amzn.to/3XfXsAO -------------------------------------------------------------------------------------------------------------------------------- If You Like tutorial like this, You Can Support Our Work In Patreon: https://www.patreon.com/c/aifuturetech #aimusic #aimusicvideo #comfyui #aimodel

Comments

coachbate.ai

if you get error about not finding hf and you are running Windows, this worked for me to install it: powershell -ExecutionPolicy ByPass -c "irm https://hf.co/cli/install.ps1 | iex"

coachbate.ai

This is great. Thank you! The way you are setting MODEL_BASE_DIR doesn't support the custom folders specified in the extra_model_paths.yaml This snippet may help resolve that and also allow people to put it anywhere they want via the extra_model_paths.yaml : # ---------------------------- # Path Configuration # ---------------------------- folder_paths.add_model_folder_path("HeartMuLa", os.path.join(folder_paths.models_dir, "HeartMuLa")) def get_model_base_dir(): paths = folder_paths.get_folder_paths("HeartMuLa") for p in paths: if os.path.exists(p): return p return paths[0] MODEL_BASE_DIR = get_model_base_dir()