The Babel Machine

In numbers

What global researchers did with the Babel Machine

500+ researchers

have run a corpus through the platform or requested access to our models.

52 countries

Researchers on six continents have run a corpus through the platform or requested our models.

250+ institutions

are home to the researchers who have used the platform or requested our models.

Thousands of classification jobs

submitted to the hosted no-code platform between September 2023 and August 2026.

4× more jobs in 2025 than in 2024

75.5 million+ rows of text

uploaded by researchers for automated coding between September 2023 and August 2026.

1.1 million+ model downloads

of our fine-tuned classifiers on Hugging Face, all time.

100+ supported languages

and 14 coding tasks: policy topics, sentiment, emotions, named entities, manifestos, illiberal frames and more.

Our users

Institutional users and affiliations

Thank you for using the Babel Machine! Due to severe and unforeseen budget cuts, the Babel Machine has temporarily suspended operations.

For inquiries or custom project requests, please contact us at our contact address.

Funding and support

This project was supported by the Ministry of Innovation and Technology NRDI Office within the RRF-2.3.1-21-2022-00004 Artificial Intelligence National Laboratory project; the V-Shift Momentum Project of the Hungarian Academy of Sciences; Miklós Sebők's Excellence project (identifier: 151324), which is funded by the Hungarian National Research, Development and Innovation Office's National Research Excellence Programme; and received additional funding from the European Union's Horizon 2020 program under grant agreement no 101008468. We also thank the Babel Machine project and HUN-REN Cloud (Héder et al. 2022; https://science-cloud.hu) for their support. We used the machine learning service of the Slices RI infrastructure (https://www.slices-ri.eu/).

No-code multilingual AI for comparative research

Political text, decoded.

The Babel Machine turns uploaded text into structured research data: CAP (Comparative Agendas Project) topics, manifesto labels, sentiment, emotions, named entities, ILLFRAMES, ONTOLISST, and entity-targeted sentiment analysis.

Choose a classifier Try the demo

14+ coding tasks

100+ supported languages

190+ institutional users worldwide

Demo

Test the classifiers live.

Upload text and explore the Babel Machine classifiers in action.

Babel Machine Demo Embedded workflow

Classifier Suite

Pick the taxonomy, upload the file, receive coded data.

Every classifier is available from the new homepage, grouped by the research decision the user is making.

Policy Topics

CAP Major (22)

Automated coding of CAP major policy topic codes.

Policy Topics

CAP Minor (214)

Automated coding of detailed CAP minor policy topic codes.

Media Coding

CAP Major + Media1

CAP major policy topic codes plus Media1 categories.

Media Coding

CAP Major + Media2

CAP major policy topics plus expanded Media2 categories.

Media Coding

CAP Minor + Media1

Fine-grained policy topics combined with Media1 codes.

Manifestos

Manifesto

Automated coding of party manifesto text.

Tone

Sentiment 3

Three-way sentiment coding for comparative text analysis.

Tone

Emotions 6

Automated coding of six emotion categories.

Tone

Emotions 9

Automated coding of nine emotion categories.

Tone

Emotions 10

Automated coding of ten emotion categories.

Frames

ILLFRAMES

Automated coding of illiberal policy frames.

Frames

ONTOLISST

Automated coding of ONTOLISST categories.

Extraction

NER

Named-Entity Recognition for people, organizations, and places.

Extraction

ABSA 3

Three-way sentiment analysis targeted at named entities inside sentences.

Coming Soon

Multi-task Upload

Multiple classification tasks with one upload.

About

poltextLAB AI Laboratory

Babel Machine is maintained by poltextLAB Artificial Intelligence Laboratory in Budapest and developed with international collaborators across comparative politics and text-as-data research.

Visit poltextLAB

Demo

Embedded workflow

Test the upload and classification flow directly in the demo section above.

Run demo

Coverage

Supported languages

The platform supports broad multilingual coverage for CAP (Comparative Agendas Project), manifesto, sentiment, emotions, ILLFRAMES, ONTOLISST, ABSA, and NER tools.

See languages

Citation

How to cite

Sebok, M., Mate, A., Ring, O., Kovacs, V., & Lehoczki, R. (2025). Leveraging Open Large Language Models for Multilingual Policy Topic Classification: The Babel Machine Approach.

DOI

Data

GDPR note

Uploaded files are processed in a secure cloud environment to return coded CSV outputs. Submission data is used only for returning results and identifying the submitting organization.

GDPR reference

Updates

Platform updates

Babel Machine continues to expand classifier coverage, language support, and research infrastructure.

Browse tools

Contact

Get in touch

Address: 35/C Marton utca, 1094 Budapest, Hungary.

Email us

Language Coverage

Built for multilingual social science, not just English demos.

Current tasks cover 100+ languages, with the broadest coverage for English, Hungarian, French, German, Polish, Czech, Slovak, Spanish, Portuguese, Italian, Dutch, and Danish.

English Hungarian French German Polish Czech Slovak Spanish Portuguese Italian Dutch Danish Norwegian Romanian Bulgarian Serbian Croatian Slovenian Swedish Finnish Estonian Latvian Lithuanian Greek Russian Hebrew Armenian Georgian Japanese Korean Icelandic

Research Backbone

Research-grade AI, wrapped in a no-code workflow.

How to cite

Sebok, M., Mate, A., Ring, O., Kovacs, V., & Lehoczki, R. (2025). Leveraging Open Large Language Models for Multilingual Policy Topic Classification: The Babel Machine Approach. Social Science Computer Review, 43(2), 295-317.

Read the article

GDPR-ready processing

Uploaded files are processed to return coded CSV outputs. Personal data associated with submission is used only for returning results and identifying the submitting organization, and users can request deletion at any time.

GDPR reference

Institutional home

Babel Machine is maintained by poltextLAB Artificial Intelligence Laboratory, a Budapest-based research lab focused on text mining and AI for social science research.

Visit poltextLAB

Users

Institutional customers and the affiliations of personal users

Researchers from 190+ institutions in 40 countries across 5 continents have used the platform and models (May 2026).

View full user overview ↗

Current sponsors and infrastructure partners.

Babel Machine is maintained by poltextLAB Artificial Intelligence Laboratory in Budapest and developed with international collaborators across comparative politics and text-as-data research.

Contact

Bring your codebook, corpus, or comparative research problem.

Email us Contact page