Ace-Step-1.5-ScragVAE

About

What Ace-Step-1.5-ScragVAE is

ScragVAE — Improved VAE Decoder for ACE-Step 1.5 A fine-tuned AutoencoderOobleck decoder with an intent to improve audio fidelity for the ACE-Step 1.5 music generation pipeline. Drop-in compatible with all existing ACE-Step DiT checkpoints. What is this? ACE-Step 1.5 uses a VAE (Variational Autoencoder) to convert between audio waveforms and the latent space that the DiT diffusion model operates in. The original VAE decoder attenuates high-frequency content, resulting in audio with reduced clarity and detail above 6kHz. ScragVAE retrains the decoder half of the VAE to better reconstruct upper harmonics, transient detail, and spectral "air" — while keeping the encoder frozen so all existing DiT models remain fully compatible. Benchmarks Objective spectral analysis comparing ScragVAE vs the original ACE-Step 1.5 VAE decoder on identical latents (same seed, same DiT output): Summary: ScragVAE preserves significantly more high-frequency content (especially 10–20kHz) and has dramatically better dynamic range, resulting in clearer vocals, crisper transients, and more natural-sounding audio. Files Usage Python / Diffusers ScragVAE is a drop-in replacement for the ACE-Step VAE. Replace the VAE checkpoint path in your pipeline: Or manually swap the decoder weights in an existing setup: acestep.cpp / HOT-Step CPP Place scragvae-BF16.gguf in your models directory alongside the other GGUF files: The engine auto-discovers all VAE GGUFs at startup. In HOT-Step CPP, select ScragVAE from the VAE Decoder dropdown in the Models & Adapters panel. For acestep.cpp's built-in web UI or API, pass "vaemodel": "scragvae-BF16.gguf" in your synth request JSON. Converting from safetensors to GGUF yourself If you need to reconvert (e.g. after further fine-tuning): Or use the converter directly: Architecture ScragVAE uses the same AutoencoderOobleck architecture as the original ACE-Step VAE — no structural changes. Only the decoder weights differ. Compatibility - ✅ All ACE-Step 1.5 DiT checkpoints (turbo, SFT, XL) - ✅ All LoRA/adapter models - ✅ Both Python (PyTorch/Diffusers) and C++ (ggml/acestep.cpp) runtimes - ✅ Encoder weights are identical — no retraining of upstream models needed Trai

Summarised from the published model card. Read the full card on the HuggingFace links below.

Specifications

What it is

How it works

How audio & music work

Variants

Sizes & precisions

Open weights ship in several sizes and precisions. One page, all the variants — pick the one that fits your GPU. VRAM figures are estimates from model size.

Using it via the API

Call it like any OpenAI endpoint

Once AxForge deploys ace-step-1-5-scragvae for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (ace-step-1-5-scragvae below is illustrative; you get the exact model name on deployment.)

$ curl -sS https://api.axforge.ai/v1/audio/transcriptions \
  -H "Authorization: Bearer $AXFORGE_API_KEY" \
  -F model="ace-step-1-5-scragvae" -F file=@audio.mp3

Details

Languages, data & research

Licence

Open weights

Sources

Weights & code

Explore

More audio & music