@newmean-ai/pii (https://github.com/newmean-ai/pii)
Copyright 2026 newmean.ai

The SDK code and the model weights (newmean-pii-nano, -small, -medium, -large; Hugging Face repositories newmean-ai/pii-nano,
-small, -medium, -large) are licensed under the Apache License 2.0 (LICENSE).

Attribution request (not a license condition): if you use the SDK or the models in a product, we would appreciate a
"Powered by newmean.ai" mention (linking to https://newmean.ai) in your documentation or an about, licenses or settings screen.

The models include or were derived from the following third-party material.

- mMiniLMv2 (Multilingual-MiniLM), Microsoft — https://github.com/microsoft/unilm/tree/master/minilm
  The four models start from mMiniLMv2-L6-H384; the vocabulary is pruned and the layers are compressed per size.
- XLM-R SentencePiece tokenizer, Facebook AI — https://github.com/facebookresearch/fairseq/tree/main/examples/xlmr
  The tokenizer files are the XLM-R tokenizer with its vocabulary pruned to the pieces used in training.
- Teacher models used only to produce training targets (not distributed): mDeBERTa-v3-base (Microsoft, MIT License) and
  XLM-RoBERTa-large (Facebook AI, MIT License), fine-tuned by newmean.ai.

Training data: synthetic text generated by newmean.ai with Claude models; no third-party dataset was used for training.
Evaluation only (not distributed): KDPII (CC BY 4.0), OpenPII 1.5M (Ai4Privacy / Ai Suisse SA, CC BY 4.0), NVIDIA
Nemotron-PII (CC BY 4.0), Gretel synthetic_pii_finance_multilingual (Apache-2.0).

--------------------------------------------------------------------------------
mMiniLMv2 — https://github.com/microsoft/unilm

The MIT License (MIT)

Copyright (c) Microsoft Corporation

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.

--------------------------------------------------------------------------------
XLM-R tokenizer (fairseq) — https://github.com/facebookresearch/fairseq

MIT License

Copyright (c) Facebook, Inc. and its affiliates.

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
