AC

ace

AceGGUF by calcuisUpdated 2025-07-14
Download on Hugging Face
Parameters3.31B
Context—
Layers—
Architecturepig
Smallest quant0.1 GB
Quick answer

ace needs about 0.6 GB of memory at Q4_K_M (a 0.1 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 27 quants, from 0.1 GB (Q4_K_M) to 13.2 GB (F32).

ACE-Step: Music generation model based on diffusion and lightweight transformer

ACE-Step is an open‑source foundation model for music generation. It combines diffusion‑based generation with a Deep Compression AutoEncoder and a lightweight linear transformer. The model supports full‑song generation, duration control, and natural language prompts.

Good at

  • Fast generation (15× faster than LLM baselines)
  • Coherent melody, harmony, and rhythm
  • Natural language music description

Limits

  • No vision support
  • Performance varies by language
  • Longer generations may lose coherence

Best for

  • Generating original music from text
  • Music remixing and style transfer
  • Voice cloning and music production tools

Who it’s for. Developers and creators looking for a lightweight, open‑source music generation model. The model is suitable for building music production tools, creative AI assistants, and downstream applications such as voice cloning.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_K_M0.1 GB 0.6 GB
Fits Get file →
Q4_00.2 GB 0.7 GB
Fits Get file →
F160.3 GB 0.9 GB
Fits Get file →
BF160.4 GB 1.0 GB
Fits Get file →
F160.4 GB 1.0 GB
Fits Get file →
F160.5 GB 1.1 GB
Fits Get file →
Q2_K5.5 GB 6.6 GB
Fits Get file →
Q3_K_S5.7 GB 6.8 GB
Fits Get file →
Q3_K_M5.7 GB 6.9 GB
Fits Get file →
Q3_K_L5.7 GB 6.9 GB
Fits Get file →
Q4_05.9 GB 7.2 GB
Fits Get file →
Q4_K_S5.9 GB 7.2 GB
Fits Get file →
Q4_K_M6.0 GB 7.2 GB
Tight Get file →
Q4_16.1 GB 7.3 GB
Tight Get file →
Q5_06.2 GB 7.5 GB
Tight Get file →
Q5_K_S6.2 GB 7.5 GB
Tight Get file →
Q5_K_M6.2 GB 7.5 GB
Tight Get file →
Q5_16.3 GB 7.6 GB
Tight Get file →
Q6_K6.5 GB 7.8 GB
Tight Get file →
F166.6 GB 7.9 GB
Tight Get file →
IQ4_XS6.7 GB 8.0 GB
Tight Get file →
IQ4_NL6.7 GB 8.0 GB
Too big Get file →
Q4_06.7 GB 8.0 GB
Too big Get file →
Q5_07.0 GB 8.3 GB
Too big Get file →
Q8_07.0 GB 8.3 GB
Too big Get file →
Q8_07.7 GB 9.1 GB
Too big Get file →
F3213.2 GB 15.3 GB
Too big Get file →

For 8 GB VRAM, use Q4_K_M (0.13 GB, 0.6 GB memory) or Q4_0 (0.22 GB, 0.7 GB memory). For 16 GB VRAM, Q5_K_M (6.25 GB, 7.5 GB memory) or Q6_K (6.49 GB, 7.8 GB memory) are suitable. Larger VRAM allows full‑precision F32 (13.22 GB, 15.3 GB memory).

What GPU do you need for ace?

These figures add the file size, an fp16 KV cache at an 8K context window and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: the best quant that fits is F16 (0.5 GB file, about 1.1 GB in use).
  • 6 GB VRAM: the best quant that fits is F16 (0.5 GB file, about 1.1 GB in use).
  • 8 GB VRAM: the best quant that fits is IQ4_NL (6.7 GB file, about 8.0 GB in use).
  • 12 GB VRAM: the best quant that fits is Q8_0 (7.7 GB file, about 9.1 GB in use).
  • 16 GB VRAM: the best quant that fits is F32 (13.2 GB file, about 15.3 GB in use).

Which ace quant should you choose?

Q4_K_M is the usual starting point for ace: it balances file size and quality well. Here it is a 0.1 GB file.

The Q8_0 file is 7.0 GB, about 55.8 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturepig
Parameters3.31B

Source

Quantized bycalcuis
FamilyAce
Licenseapache-2.0
Downloads (HF)8,412
Last updated2025-07-14

Frequently asked questions about ace

How much VRAM does ace need?

About 0.6 GB at Q4_K_M with an 8K context window. That is the 0.1 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run ace on an 8 GB GPU?

Yes, with IQ4_NL: the 6.7 GB file needs about 8.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of ace GGUF?

It depends on the quant: Q4_K_M is 0.1 GB, Q4_0 is 0.2 GB, F16 is 0.3 GB, BF16 is 0.4 GB, F16 is 0.4 GB, F16 is 0.5 GB and 21 more.

Where can I download ace GGUF?

The files are on Hugging Face in the calcuis/ace-gguf repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does ace use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run ACE-Step on 8 GB VRAM?

Yes, use the Q4_K_M or Q4_0 quantized files which require about 0.6–0.7 GB memory at 8k context.

Does ACE-Step support image inputs?

No, the model does not support vision inputs.

What license does ACE-Step use?

ACE-Step is released under the Apache 2.0 license.

Similar models

Browse all →

More from calcuis

All by calcuis →

More Ace models

All Ace →

Other tiny models

All tiny models →