Author:

OS:

Linux

License:

MPL

Tags:

Taggers, Tools

NameTag 3

Introduction
Current Release
License
Models
Results at a Glance
Acknowledgements
- 6.1. Publications
Contact

1. Introduction

NameTag 3 is an open-source tool for both flat and nested named entity recognition (NER). NameTag 3 identifies proper names in text and classifies them into a set of predefined categories, such as names of persons, locations, organizations, etc.

NameTag 3 achieves state-of-the-art performance on 33 test datasets in 23 languages: Cebuano, Chinese, Croatian, Czech, Danish, English, German, Greek, Hebrew, Indonesian, Japanese, Korean, Norwegian Bokmål, Norwegian Nynorsk, Portuguese, Romanian, Russian, Serbian, Slovak, Slovenian, Swedish, Tagalog, and Ukrainian. It also delivers competitive results on Arabic, Dutch, Maghrebi Arabic, and Spanish, as of May 2026.

NameTag 3 offers trained models for the following 20 languages: Arabic, Chinese, Croatian, Czech, Danish, Dutch, English, German, Greek, Hebrew, Maghrebi Arabic, Norwegian Bokmål, Norwegian Nynorsk, Portuguese, Serbian, Slovak, Slovenian, Spanish, Swedish, and Ukrainian.

NameTag is available in the following versions:

NameTag 3:
- NER tool for both flat and nested named entity recognition,
- fine-tuned LLM with either a softmax classification head for flat NER or a seq2seq decoding head for nested NER,
- NameTag 3 is available as a commandline Python script, as a web application, and NameTag REST Web Service hosted by LINDAT/CLARIN.
NameTag 2:
- NER tool,
- uses frozen BERT contextualized word embeddings with LSTM contextualization layer,
- recognizes both flat and nested named entities of arbitrary depth, depending on the trained model,
- NameTag 2 is available as a web application NameTag Web Application, and NameTag REST Web Service hosted by LINDAT/CLARIN.
NameTag 1:
- fast and robust NER tool,
- feedforward neural network,
- recognizes flat named entities,
- available as commandline tool, C++ source code, with bindings in Perl, Python and C#,
- NameTag 1 is available as a web application NameTag Web Application, and NameTag REST Web Service hosted by LINDAT/CLARIN.

NameTag 3 is a free software under Mozilla Public License 2.0, and the linguistic models are free for non-commercial use and distributed under CC BY-NC-SA license, although for some models the original data used to create the model may impose additional licensing conditions. NameTag is versioned using Semantic Versioning.

2. Current Release

NameTag 3 can be used either as a commandline tool or by requesting the NameTag webservice:

LINDAT/CLARIN hosts the NameTag Web Application.
LINDAT/CLARIN also hosts the NameTag REST Web Service.

NameTag 3 source code can be found at GitHub.

3. License

NameTag 3 is free software distributed under the Mozilla Public License 2.0.

Unless stated otherwise, the associated linguistic models and data are available for non-commercial use under the CC BY-NC-SA 4.0 license. Some models may be subject to additional licensing conditions imposed by the original data used to create them. Please consult the NameTag 3 Models webpage before using one of the models.

We are currently developing models trained exclusively on data available under more permissive licenses. Our aim is to make commercial licensing available for these models in the future. Commercial licenses are not currently available unless explicitly stated otherwise.

NameTag is versioned according to Semantic Versioning.

If you use NameTag in scientific work, please cite Straková & Straka (2025).

4. Models

The individual models are described on the NameTag 3 Models webpage and are distributed via the LINDAT repository.

5. Results at a Glance

Model	Multi	Multi	Czech CNEC 2.0	NameTag 3
Version	`260521`	`250203`	`240830`	All
Languages trained	20	17	1	20
Languages evaluated	27	20	1	27
Languages SOTA	23	15	1	23
Datasets trained	25	21	1	26
Datasets evaluated	39	28	1	40
Datasets SOTA	31	20	1	33

6. Acknowledgements

Acknowledgements for the individual language models are listed in NameTag 3 Models page.

This work has been supported by the MŠMT OP JAK program, project No. CZ.02.01.01/00/22_008/0004605 and by the Grant Agency of the Czech Republic under the EXPRO program as project “LUSyD” (project No. GX20-16819X). The work described herein has also been using data provided by the LINDAT/CLARIAH-CZ Research Infrastructure, supported by the Ministry of Education, Youth and Sports of the Czech Republic (Project No. LM2023062).

6.1. Publications

Jana Straková and Milan Straka. 2025. NameTag 3: A Tool and a Service for Multilingual/Multitagset NER . In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 3: System Demonstrations), pages 31–39, Vienna, Austria. Association for Computational Linguistics.

@inproceedings{strakova-straka-2025-nametag,
    title = "{N}ame{T}ag 3: A Tool and a Service for Multilingual/Multitagset {NER}",
    author = "Strakov{\'a}, Jana  and
      Straka, Milan",
    editor = "Mishra, Pushkar  and
      Muresan, Smaranda  and
      Yu, Tao",
    booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 3: System Demonstrations)",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2025.acl-demo.4/",
    pages = "31--39",
    ISBN = "979-8-89176-253-4",
}

7. Contact

Authors:

NameTag website.

Screenshot:

Institute of Formal and Applied Linguistics

Charles University, Czech Republic
Faculty of Mathematics and Physics

Search form