ONNX Model Zoo

zukky/Ace-Step1.5-ONNX-DLL

By zukky · ↓ 0 · ♥ 0 · Updated 2026-02-15

What's in this repo

This page lists the files in zukky/Ace-Step1.5-ONNX-DLL, including audio_detokenizer.onnx, audio_tokenizer.onnx, condition_encoder.onnx and more.

  • 21 ONNX models (.onnx files)
  • 67 files, including 21 .onnx, 17 .data, 8 .json, 4 .cs, 3 .md, 3 .cache
  • Last updated 2026-02-15

ONNX files

File Size Download
onnx/audio_detokenizer.onnx — Download
onnx/audio_tokenizer.onnx — Download
onnx/condition_encoder.onnx — Download
onnx/dit_decode_kv.onnx — Download
onnx/dit_decoder.onnx — Download
onnx/dit_decoder_probe.onnx — Download
onnx/dit_decoder_sublayer_probe.onnx — Download
onnx/dit_prefill_kv.onnx — Download
onnx/lm_0p6/decode/lm_0p6_decode.onnx — Download
onnx/lm_0p6/prefill/lm_0p6_prefill.onnx — Download

Showing 10 of 21 .onnx files — see the full file list on the right.

How to use

These ONNX files can be loaded with ONNX Runtime. To inspect a model graph, upload the file to the Netron viewer. Download links point to the original files on Hugging Face.

README

--- license: mit library_name: onnxruntime pipeline_tag: text-to-audio tags:

  • text-to-audio
  • music-generation
  • onnx
  • rust
  • ffi
  • windows

ACE-Step 1.5 ONNX DLL (Windows)

この配布物は、Hugging Face の ACE-Step/Ace-Step1.5 を ONNX 化して DLL/FFI で利用できるようにしたものです。

このディレクトリは acestep_runtime.dll の配布物です。

Contents

  • bin/acestep_runtime.dll
  • include/acestep_runtime.h
  • docs/dll_api.md
  • examples/ffi/python/*.py
  • examples/ffi/csharp/*
  • onnx/*（.onnx + .onnx.data）
  • manifest_release.json
  • LICENSE
  • THIRD_PARTY_LICENSES_RUST.md

License (Rust / DLL)

  • acestep_runtime.dll 本体コード: MIT（LICENSE）
  • Rust依存クレートのライセンス一覧: THIRD_PARTY_LICENSES_RUST.md
  • ONNX Runtime はホスト側ランタイムを利用（ライセンスは各配布物を参照）

Prerequisites

  • OS: Windows x64
  • Python 3.10+
  • Python packages: onnxruntime, numpy, soundfile
pip install onnxruntime numpy soundfile

Verified Commands

1) DLL FFI smoke test

$env:ACESTEP_RUNTIME_DLL = '.\bin\acestep_runtime.dll'
python .\examples\ffi\python\regression.py

Expected:

  • python ffi regression: PASS

2) Python + ONNX Runtime + DLL demo

このデモは runtime_py_ort モジュールを参照するため、PYTHONPATH にリポジトリルートを設定してください。

$env:PYTHONPATH = '.'
python .\examples\ffi\python\onnxruntime_dll_demo.py `
  --case fixtures/cases/text2music_01.json `
  --inputs-npz fixtures/tensors/text2music_01.npz `
  --onnx-dir .\onnx `
  --dll .\bin\acestep_runtime.dll `
  --provider cpu `
  --out-wav .\reports\python_ort_dll.wav `
  --out-npz .\reports\python_ort_dll.npz

Expected:

  • Wrote wav: .\reports\python_ort_dll.wav
  • Wrote npz: .\reports\python_ort_dll.npz

Notes

  • onnx/*.onnx.data を削除するとモデルが壊れます。
  • GPUで実行する場合は --provider cuda と CUDA 対応 ONNX Runtime を使用してください。

Hash Check (optional)

Get-FileHash .\bin\acestep_runtime.dll -Algorithm SHA256

Model info

text-to-audio onnxruntime
onnxruntime onnx text-to-audio music-generation rust ffi windows license:mit region:us
↓ 0 ♥ 0

Created: 2026-02-15

Updated: 2026-02-15

View on Hugging Face

Files (67)

.gitattributes
LICENSE
README.md
THIRD_PARTY_LICENSES_RUST.md
bin/acestep_runtime.dll
docs/dll_api.md
examples/ffi/csharp/Program.cs
examples/ffi/csharp/ffi_demo.csproj
examples/ffi/csharp/obj/Debug/net8.0/.NETCoreApp,Version=v8.0.AssemblyAttributes.cs
examples/ffi/csharp/obj/Debug/net8.0/ffi_demo.AssemblyInfo.cs
examples/ffi/csharp/obj/Debug/net8.0/ffi_demo.AssemblyInfoInputs.cache
examples/ffi/csharp/obj/Debug/net8.0/ffi_demo.GeneratedMSBuildEditorConfig.editorconfig
examples/ffi/csharp/obj/Debug/net8.0/ffi_demo.GlobalUsings.g.cs
examples/ffi/csharp/obj/Debug/net8.0/ffi_demo.assets.cache
examples/ffi/csharp/obj/ffi_demo.csproj.nuget.dgspec.json
examples/ffi/csharp/obj/ffi_demo.csproj.nuget.g.props
examples/ffi/csharp/obj/ffi_demo.csproj.nuget.g.targets
examples/ffi/csharp/obj/project.assets.json
examples/ffi/csharp/obj/project.nuget.cache
examples/ffi/python/demo.py
examples/ffi/python/onnxruntime_dll_demo.py
examples/ffi/python/regression.py
include/acestep_runtime.h
manifest_release.json
onnx/audio_detokenizer.onnx ONNX
onnx/audio_tokenizer.onnx ONNX
onnx/condition_encoder.onnx ONNX
onnx/condition_encoder.onnx.data
onnx/dit_decode_kv.onnx ONNX
onnx/dit_decode_kv.onnx.data
onnx/dit_decoder.onnx ONNX
onnx/dit_decoder.onnx.data
onnx/dit_decoder_probe.onnx ONNX
onnx/dit_decoder_probe.onnx.data
onnx/dit_decoder_sublayer_probe.onnx ONNX
onnx/dit_decoder_sublayer_probe.onnx.data
onnx/dit_prefill_kv.onnx ONNX
onnx/dit_prefill_kv.onnx.data
onnx/io_contract_core.json
onnx/io_contract_lm_0p6.json
onnx/io_contract_lm_1p7.json
onnx/io_contract_qwen3_embedding_0p6.json
onnx/lm_0p6/decode/lm_0p6_decode.onnx ONNX
onnx/lm_0p6/decode/lm_0p6_decode.onnx.data
onnx/lm_0p6/prefill/lm_0p6_prefill.onnx ONNX
onnx/lm_0p6/prefill/lm_0p6_prefill.onnx.data
onnx/lm_0p6_decode.onnx ONNX
onnx/lm_0p6_decode.onnx.data
onnx/lm_0p6_prefill.onnx ONNX
onnx/lm_0p6_prefill.onnx.data
onnx/lm_1p7/decode/lm_1p7_decode.onnx ONNX
onnx/lm_1p7/decode/lm_1p7_decode.onnx.data
onnx/lm_1p7/lm_1p7_decode.onnx ONNX
onnx/lm_1p7/lm_1p7_decode.onnx.data
onnx/lm_1p7/lm_1p7_prefill.onnx ONNX
onnx/lm_1p7/lm_1p7_prefill.onnx.data
onnx/lm_1p7/prefill/lm_1p7_prefill.onnx ONNX
onnx/lm_1p7/prefill/lm_1p7_prefill.onnx.data
onnx/lm_1p7_decode.onnx ONNX
onnx/lm_1p7_decode.onnx.data
onnx/lm_1p7_prefill.onnx ONNX
onnx/lm_1p7_prefill.onnx.data
onnx/manifest.json
onnx/qwen3_embedding_0p6.onnx ONNX
onnx/qwen3_embedding_0p6.onnx.data
onnx/qwen3_token_embedding_0p6.onnx ONNX
onnx/vae_decoder.onnx ONNX

Related models

More from zukky

More text-to-audio models

Browse all text-to-audio models

Popular models