TAAFT
Free mode
100% free
Freemium
Free Trial
Deals
Companies
StepFun
Models
ACE Step v1.5

ACE Step v1.5

By StepFun
Model family: Step
ACE-STEP v1.5 upgrades the base ACE-STEP architecture into a production grade, locally runnable music generator. A language model first expands text prompts into a structured blueprint of lyrics, chords and arrangement, which then conditions a DiT based audio generator. The model uses intrinsic reinforcement learning for alignment and supports style control, vocal covers and editing workflows like repaint and vocal to BGM, while generating full length music in a few seconds on A100 or single gaming GPUs.
New Audio Gen 4
Released: February 1, 2026

Overview

ACE-STEP v1.5 is an open source, super fast music foundation model that uses a hybrid language model plus diffusion transformer pipeline to turn short prompts into multi minute songs, running on consumer GPUs with under 4 GB VRAM

About StepFun

StepFun AI is a creative AI assistant platform that offers chat, search, coding help, and multimodal generation (images, audio, video), plus research models like Step3 and NextStep-1.

Website: stepfun.ai
View Company Profile

Other models from this family

  • Gen 7Step
    Step 3.5 Flash
    By StepFun
    Step 3.5 Flash is StepFun’s open source sparse MoE LLM, with 196B total parameters but 11B active per token, tuned for fast agentic reasoning, coding and long context work while staying efficient.
    NewText
    Released 16d ago
  • Gen 4Step
    Step-Audio-EditX
    By StepFun
    Step-Audio-EditX is a text-guided audio editor for speech, music, and effects. Describe the change and it performs precise, time-aligned edits with low latency.
    Audio
    Released 3mo ago

Tools using ACE Step v1.5

No tools found for this model yet.

Related Models

  • Gen 4
    LuxTTS
    ZipVoice-based voice cloning TTS that generates 48 kHz speech at up to 150x real time, fitting in about 1 GB VRAM for local, high quality synthesis
    NewAudio
    Released 16d ago
  • Gen 4
    Scribe v2 Realtime
    Scribe v2 Realtime is a low-latency speech model for live transcription and captioning. It streams partial and final text with stable timestamps, optional speaker labels, multilingual recognition, and clean JSON events for apps that need instant, accurate audio-to-text.
    Audio
    Released 3mo ago
  • Gen 4
    Kyutai Pocket TTS 1.6B
    Kyutai TTS 1.6B is Kyutai's open-source streaming text-to-speech model for English and French, using delayed streams modeling to start speaking before the full text is read, enabling ultra-low-latency, high-quality voices for assistants and real-time apps.
    NewAudio
    Released 1mo ago
Last updated: February 4, 2026
0 AIs selected
Clear selection
#
Name
Task