Downloads · 30 days
0
Ace-Step-v1.5-ScragVAE-mlx
LumeryApp/Ace-Step-v1.5-ScragVAE-mlx
Ace-Step-v1.5-ScragVAE-mlx is a text-to-audio model from LumeryApp. Use it for the text-to-audio task on the model card, and read the license before you ship it in a product. It is set up for mlx. The card lists the license as mit.
MLX-format conversion of ScragVAE — a community-finetuned drop-in replacement for the AceStep 1.5 audio VAE. It shares the AutoencoderOobleck architecture and config schema, so it substitutes directly for the bundled…
Access
Public
Updated Jul 23, 2026
Repo size
—
Likes
0
Public
Repo makeup
Click a slice to open those files.
Other2.4 KB · 70%
At a glance
- Task
- Text-to-Audio
- Library
- mlx
- License
- mit
- Access
- Public
- Created
- Jul 23, 2026
- Updated
- Jul 23, 2026
- SHA
- d7e78aca
Base models
Model card
From the Hugging Face model README
ScragVAE (Ace-Step 1.5) — MLX (LumeryApp)
MLX-format conversion of ScragVAE — a community-finetuned drop-in replacement for the
AceStep 1.5 audio VAE. It shares the AutoencoderOobleck architecture and config schema,
so it substitutes directly for the bundled decoder. Lumery uses it as the default
production decode chain for AceStep music.
- Format: MLX, BF16 (
model.safetensors+config.json). - Upstream:
scragnog/Ace-Step-1.5-ScragVAE, © @scragnog, licensed MIT. - Modifications: conversion of the original weights to MLX safetensors. No retraining.
Provenance & attribution
Unofficial community conversion — not affiliated with or endorsed by @scragnog. See
NOTICE for the MIT attribution.
License
MIT, inherited from upstream. See NOTICE.
- Task
- Text-to-Audio
- Library
- mlx
- License
- mit
- Created
- Jul 23, 2026
- Updated
- Jul 23, 2026