Skip to content
SPECIFICTION
Get listed

HeartMuLa

heartmula.github.io

WebSelf-hosted
Visit heartmula.github.io
Spec score 7.2 #112 of 123 in AI music generators

Spec sheet

Product
HeartMuLa is an open-source family of music foundation models for music understanding and generation.Read 12 Oct 2026 · heartmula.github.io
Supported platforms
webRead 21 Sep 2026 · heartmula.github.io
Integrations
The repository links to a community-created ComfyUI custom node for HeartMuLa.Read 12 Oct 2026 · github.com
Support
The maintainers list [email protected] and invite users to join their Discord or WeChat group.Read 12 Oct 2026 · github.com
License
The repository and related model weights are stated to use the Apache 2.0 license.Read 12 Oct 2026 · github.com
Audio codec
HeartCodec is described as a 12.5 Hz music codec with high reconstruction fidelity.Read 12 Oct 2026 · github.com
Audio export formats
mp3Read 21 Sep 2026 · heartmula.github.io
Audio-text alignment
HeartCLAP aligns audio and text in a shared embedding space for music descriptions and cross-modal retrieval.Read 12 Oct 2026 · heartmula.github.io
Audio-text retrieval
HeartCLAP aligns audio and text in a shared embedding space for music descriptions and cross-modal retrieval.Read 12 Oct 2026 · heartmula.github.io
Commercial use
YesRead 21 Sep 2026 · heartmula.github.io
Downloads
The repository provides pretrained checkpoints through Hugging Face and ModelScope for local deployment.Read 12 Oct 2026 · github.com
Generation controls
The models support multi-conditional creation using style descriptions, lyrics, and reference audio; the site also describes section-level style control and short music generation for videos.Read 12 Oct 2026 · heartmula.github.io
Hardware
The setup instructions use CUDA devices and describe splitting model parameters across multiple GPUs or enabling lazy loading to save GPU memory.Read 12 Oct 2026 · github.com
Integration
The repository links to a community-created ComfyUI custom node for HeartMuLa.Read 12 Oct 2026 · github.com
Known limitation
The README says the 7B version was not yet released when its usage instructions were written.Read 12 Oct 2026 · github.com
Limit
The repository documents a default maximum generated audio length of 240,000 milliseconds.Read 12 Oct 2026 · github.com
Local deployment
The repository provides instructions to install HeartMuLa locally with Python and run music generation from downloaded model checkpoints.Read 12 Oct 2026 · github.com
Model components
The family includes HeartMuLa, HeartCodec, HeartTranscriptor, and HeartCLAP.Read 12 Oct 2026 · github.com
Model family
The family includes HeartMuLa, HeartCodec, HeartTranscriptor, and HeartCLAP.Read 12 Oct 2026 · github.com
Music codec
HeartCodec is described as a 12.5 Hz music codec with high reconstruction fidelity.Read 12 Oct 2026 · github.com
Online demos
The maintainers say they launched online demo spaces on Hugging Face and ModelScope.Read 12 Oct 2026 · github.com
Requirements
The local deployment instructions recommend Python 3.10 and document CUDA device settings for inference.Read 12 Oct 2026 · github.com
Song generation
HeartMuLa generates music conditioned on lyrics and tags, with multilingual support.Read 12 Oct 2026 · github.com
Transcription
HeartTranscriptor is a Whisper-based model tuned for lyric transcription.Read 12 Oct 2026 · github.com
Vocal generation
YesRead 21 Sep 2026 · heartmula.github.io

Where Specifiction places it

Alternatives to HeartMuLaAll alternatives →