RE-USE MLX + Core ML (unofficial mirror for Nogmi)

This repository is an unofficial, unmodified mirror kept so the Nogmi app can download the speech-enhancement model from a pinned, stable location. It is not affiliated with NVIDIA or the ReuseKit authors.

  • Source repository: AILogDev/RE-USE-MLX-CoreML
  • Source revision: 4b7e50680ba34f10865cab7e24054262596f5ab4
  • Mirrored model assets: 38 files, 54,206,689 bytes. Paths and bytes are identical to the source revision; sizes and SHA-256 hashes are listed in ModelCatalog.json and manifest.json.
  • Changed: only this README.md, which replaces the source model card.
  • Not mirrored: ReuseKit.zip (the SDK source archive). Get it from the source repository.

Upstream model and license

The model materials are a conversion of NVIDIA's nvidia/NVIDIA-NemotronLabs-RE-USE (revision 31f82f2efad1095034492f236e8dac9239ba713b), distributed under the OpenMDW License Agreement, version 1.1 (OpenMDW-1.1). Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved. The full license text and notices are in LICENSE, kept unchanged from the source.

ReuseKit notice

The model pack format and the MLX / Core ML conversion come from ReuseKit / RE-USE Lab.

ReuseKit / RE-USE Lab Copyright 2026 ReuseKit contributors

Licensed under the Apache License, Version 2.0. Model weights and third-party components retain their own license terms.

The ReuseKit SDK code is Apache-2.0; its full LICENSE, NOTICE and licenses/ are in ReuseKit.zip in the source repository.

Downloads last month
119
MLX
Hardware compatibility
Log In to add your hardware

Quantized

Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for okdohyuk/RE-USE-MLX-CoreML

Quantized
(8)
this model