Originally posted: <a target="_blank" rel="ugc" href="https://huggingface.co/ACE-Step/ACE-Step-v1-3.5B">https://huggingface.co/ACE-Step/ACE-Step-v1-3.5B</a><a target="_blank" rel="ugc" href="https://github.com/ace-step/ACE-Step">https://github.com/ace-step/ACE-Step</a><h2 id="model-description">Model Description</h2>ACE-Step is a novel open-source foundation model for music generation that overcomes key limitations of existing approaches through a holistic architectural design. It integrates diffusion-based generation with Sana's Deep Compression AutoEncoder (DCAE) and a lightweight linear transformer, achieving state-of-the-art performance in generation speed, musical coherence, and controllability.Key Features:<ul><li>15× faster than LLM-based baselines (20s for 4-minute music on A100)</li><li>Superior musical coherence across melody, harmony, and rhythm</li><li>full-song generation, duration control and accepts natural language descriptions</li></ul><h2 id="uses">Uses</h2><h3 id="direct-use">Direct Use</h3>ACE-Step can be used for:<ul><li>Generating original music from text descriptions</li><li>Music remixing and style transfer</li><li>edit song lyrics</li></ul><h3 id="downstream-use">Downstream Use</h3>The model serves as a foundation for:<ul><li>Voice cloning applications</li><li>Specialized music generation (rap, jazz, etc.)</li><li>Music production tools</li><li>Creative AI assistants</li></ul>

ACE-Step Audio Gen

aceStepAudioGen_v15XLBase.safetensors

v1.5 XL Base

TEnc qwen_4b_ace15

TEnc qwen_1.7b_ace15

TEnc qwen_0.6b_ace15

v1.5 XL Turbo

v1.5 XL SFT

v1.5 Turbo

v1.5 Base

Files with SHA256: 56bf816fc9a69a5f45635e867b2ad742e1e648eb51fadb7d124cb8332d2e0940