ACE-Step v1.5 Base
ACE-Step v1.5 Base is a Runware-hosted music model built for creator workflows, with stronger fidelity, more reliable stylistic consistency, and prompt-driven genre control for full-song generation.
Added Apr 3, 2026
Details
- Approx. Price
- $0.009 per audio
- Model type
- music
- Category
- music
- Preview examples
- 0
Related audio models
Compare ACE-Step v1.5 Base with similar models from the same provider or model family.
ACE-Step 1.5 XL Turbo
ACE-Step-v1.5-XL-TurboACE-Step 1.5 XL Turbo uses a 4B music generation architecture with distilled 8-step inference. Create music from text, transform a reference track into a cover, replace a segment, or extend audio before or after the source. Supports optional lyrics, musical key, tempo, and vocal language.
ACE-Step v1.5 Turbo
ACE-Step-v1.5-TurboACE-Step v1.5 Turbo is the faster, lower-cost Runware variant for full-song generation, with broad genre coverage, improved stylistic consistency, and text-guided music creation for creator workflows.
ACE-Step 1.5
ACE-Step-1.5ACE-Step 1.5 composes complete songs from text descriptions. Guide genre, mood, and structure with style tags and required custom lyrics. Generates up to 4 minutes of multi-track audio with vocals.
ACE-Step
Wavespeed-ACE-StepACE-Step composes complete songs from text descriptions. Guide genre, mood, and structure with style tags and optional custom lyrics. Generates up to 4 minutes of multi-track audio with vocals.
ElevenLabs Scribe v3
elevenlabs/scribe-v3ElevenLabs Scribe v3 transcribes recorded speech with automatic language detection, word timestamps, optional speaker labels, and audio event tagging. Choose digits or words for numbers, add secondary language hints, or provide up to 100 keyterms. Supports MP3, WAV, OGG, M4A, and AAC recordings up to 60 minutes, with direct uploads up to 4 MB or audio URLs up to 25 MB. Pricing is $0.01 per input minute, or $0.012 with keyterms, with a 10-second minimum and proportional billing thereafter.
Voxtral Small 24B
mistralai/voxtral-small-24b-2507Mistral AI’s original Voxtral Small 24B transcribes recorded speech in English, French, German, Spanish, Italian, Portuguese, Dutch, and Hindi, with automatic language detection. Add a context prompt for names or terminology. Accepts MP3, WAV, FLAC, OGG, and M4A recordings up to 10 minutes, with direct uploads up to 4 MB or audio URLs up to 25 MB. Returns a completed JSON transcript without timestamps, speaker diarization, or translation.