NVIDIA model and upstream code:
Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
OpenMDW-1.1. Full license and third-party notices: web/MODEL-LICENSE.txt.
Source: https://huggingface.co/nvidia/NVIDIA-NemotronLabs-RE-USE
Revision: 31f82f2efad1095034492f236e8dac9239ba713b

WebGPU inference engine:
Copyright (c) 2026 geronimo
MIT. Full license: web/ENGINE-LICENSE.txt.
Source: https://github.com/geronimi73/re-use-demo
Revision: 0d54dca074597b111a8e8c78f6067a6fa0231f75
The five files in web/engine/ are unmodified copies of src/engine/ at this revision.

The static UI, local server, and verification scripts were created for this local experiment.
The model author's and engine author's trademarks do not imply endorsement of this prototype.

Deployment note (2026-11-02):
This public copy is served at /re-use/. In the original notice above,
web/MODEL-LICENSE.txt, web/ENGINE-LICENSE.txt, and web/engine/ correspond to
MODEL-LICENSE.txt, ENGINE-LICENSE.txt, and engine/ relative to this page.
The original notice above is preserved verbatim. Both license files and all
five engine files are unmodified. The original F32 safetensors model is
unchanged: only its transport is split into two same-origin binary files.
The browser reconstructs the complete bytes and verifies SHA-256 before use:
06badaf4029b297f97f7fe48d62990b48d5482608210df577354e0d4e0ed70e1
The UI and loader have been adapted for this public deployment.