This is the BF16 version of ACE-Step/acestep-v15-xl-base — the XL (4B) Base variant of ACE-Step 1.5. This BF16 conversion reduces memory usage while maintaining near-identical quality to the original model. It is a full-quality base model that generates high-quality audio in 50 steps with CFG, offering high diversity and support for all tasks (extract, lego, complete).

XL Architecture

Parameter	Value
DiT Decoder hidden_size	2560
DiT Decoder layers	32
DiT Decoder attention heads	32
Encoder hidden_size	2048
Encoder layers	8
Total params	~4B
Weights size (bf16)	~7.5 GB
Inference steps	50 (with CFG)

GPU Requirements

VRAM	Support
≥8 GB	With CPU offload + INT8 quantization
≥12 GB	With CPU offload
≥16 GB	Without offload (recommended)
≥20 GB	Full quality (XL + 4B LM)

All LM models (0.6B / 1.7B / 4B) are fully compatible with XL.

Key Features

💰 Commercial-Ready: Trained on legally compliant datasets. Generated music can be used for commercial purposes.
📚 Safe Training Data: Licensed music, royalty-free/public domain, and synthetic (MIDI-to-Audio) data.
🎯 High Diversity: Base model offers the highest diversity among XL variants.
🔧 All Tasks: Supports extract, lego, and complete tasks.
🧠 BF16 Precision: Converted to BF16 for reduced VRAM usage and faster inference, with negligible quality loss.

Quick Start

bash

# Install ACE-Step
git clone https://github.com/ace-step/ACE-Step-1.5.git
cd ACE-Step-1.5
pip install -e .

# Download this model
huggingface-cli download marcorez8/acestep-v15-xl-base-bf16 --local-dir ./checkpoints/acestep-v15-xl-base-bf16

# Run with Gradio UI
python acestep --config-path acestep-v15-xl-base-bf16

Model Zoo

XL (4B) DiT Models

DiT Model	CFG	Steps	Quality	Diversity	Tasks	Hugging Face	ModelScope
`acestep-v15-xl-base`	✅	50	High	High	All (extract, lego, complete)	Link	Link
`acestep-v15-xl-base-bf16`	✅	50	High	High	All (extract, lego, complete)	This repo	—
`acestep-v15-xl-sft`	✅	50	Very High	Medium	Standard	Link	Link
`acestep-v15-xl-turbo`	❌	8	Very High	Medium	Standard	Link	Link
`acestep-v15-xl-turbo-bf16`	❌	8	Very High	Medium	Standard	Link	—

LM Models (all compatible with XL)

LM Model	Params	Audio Understanding	Composition	Hugging Face	ModelScope
`acestep-5Hz-lm-0.6B`	0.6B	Medium	Medium	Link	Link
`acestep-5Hz-lm-1.7B`	1.7B	Medium	Medium	Included in main	Included in main
`acestep-5Hz-lm-4B`	4B	Strong	Strong	Link	Link

Acknowledgements

This project is co-led by ACE Studio and StepFun. The BF16 conversion was done by marcorez8 to make the model more accessible to the community.

Citation

BibTeX

@misc{gong2026acestep,
  title={ACE-Step 1.5: Pushing the Boundaries of Open-Source Music Generation},
  author={Junmin Gong, Yulin Song, Wenxiao Zhao, Sen Wang, Shengyuan Xu, Jing Guo},
  howpublished={\url{https://github.com/ace-step/ACE-Step-1.5}},
  year={2026},
  note={GitHub repository}
}