EuroLLM-9B: Technical Report Article Swipe

PDF

Pedro Henrique Martins , João Alves , Patrick Fernandes , Ricardo Rei , M. Amin Farajian , Mateusz Klimaszewski , Duarte M. Alves , José P. Pombal , Nicolas Boizard , Manuel Faysse , Pierre Colombo , François Yvon , Barry Haddow , José G. C. de Souza , Alexandra Birch , André F. T. Martins ·

YOU? · · 2025 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2506.04079

This report presents EuroLLM-9B, a large language model trained from scratch to support the needs of European citizens by covering all 24 official European Union languages and 11 additional languages. EuroLLM addresses the issue of European languages being underrepresented and underserved in existing open large language models. We provide a comprehensive overview of EuroLLM-9B's development, including tokenizer design, architectural specifications, data filtering, and training procedures. We describe the pre-training data collection and filtering pipeline, including the creation of EuroFilter, an AI-based multilingual filter, as well as the design of EuroBlocks-Synthetic, a novel synthetic dataset for post-training that enhances language coverage for European languages. Evaluation results demonstrate EuroLLM-9B's competitive performance on multilingual benchmarks and machine translation tasks, establishing it as the leading open European-made LLM of its size. To support open research and adoption, we release all major components of this work, including the base and instruction-tuned models, the EuroFilter classifier, and the synthetic post-training dataset.

Related Topics

Minority Report (Film)

World Happiness Report

Drudge Report

Weather Report

The Report (2019 Film)

Technical Ecstasy

The Majority Report With Sam Seder

Technical University Of Munich

U.S. News & World Report Best Colleges Ranking

Xbox 360 Technical Problems

Technical (Vehicle)

Playstation 3 Technical Specifications

Concepts

No concepts available.

Metadata

Type: preprint
Language: en
Landing Page: http://arxiv.org/abs/2506.04079
PDF: https://arxiv.org/pdf/2506.04079
OA Status: green
OpenAlex ID: https://openalex.org/W4416075033

All OpenAlex metadata

Raw OpenAlex JSON

OpenAlex ID: https://openalex.org/W4416075033

Canonical identifier for this work in OpenAlex
DOI: https://doi.org/10.48550/arxiv.2506.04079

Digital Object Identifier
Title: EuroLLM-9B: Technical Report

Work title
Type: preprint

OpenAlex work type
Language: en

Primary language
Publication year: 2025

Year of publication
Publication date: 2025-06-04

Full publication date if available
Authors: Pedro Henrique Martins, João Alves, Patrick Fernandes, Ricardo Rei, M. Amin Farajian, Mateusz Klimaszewski, Duarte M. Alves, José P. Pombal, Nicolas Boizard, Manuel Faysse, Pierre Colombo, François Yvon, Barry Haddow, José G. C. de Souza, Alexandra Birch, André F. T. Martins

List of authors in order
Landing page: https://arxiv.org/abs/2506.04079

Publisher landing page
PDF URL: https://arxiv.org/pdf/2506.04079

Direct link to full text PDF
Open access: Yes

Whether a free full text is available
OA status: green

Open access status per OpenAlex
OA URL: https://arxiv.org/pdf/2506.04079

Direct OA link when available
Cited by: 0

Total citation count in OpenAlex

Full payload

id	https://openalex.org/W4416075033
doi	https://doi.org/10.48550/arxiv.2506.04079
ids.doi	https://doi.org/10.48550/arxiv.2506.04079
ids.openalex	https://openalex.org/W4416075033
fwci	0.0
type	preprint
title	EuroLLM-9B: Technical Report
biblio.issue
biblio.volume
biblio.last_page
biblio.first_page
is_xpac	False
apc_list
apc_paid
language	en
locations[0].id	pmh:oai:arXiv.org:2506.04079
locations[0].is_oa	True
locations[0].source.id	https://openalex.org/S4306400194
locations[0].source.issn
locations[0].source.type	repository
locations[0].source.is_oa	True
locations[0].source.issn_l
locations[0].source.is_core	False
locations[0].source.is_in_doaj	False
locations[0].source.display_name	arXiv (Cornell University)
locations[0].source.host_organization	https://openalex.org/I205783295
locations[0].source.host_organization_name	Cornell University
locations[0].source.host_organization_lineage	https://openalex.org/I205783295
locations[0].license	cc-by
locations[0].pdf_url	https://arxiv.org/pdf/2506.04079
locations[0].version	submittedVersion
locations[0].raw_type	text
locations[0].license_id	https://openalex.org/licenses/cc-by
locations[0].is_accepted	False
locations[0].is_published	False
locations[0].raw_source_name
locations[0].landing_page_url	http://arxiv.org/abs/2506.04079
locations[1].id	doi:10.48550/arxiv.2506.04079
locations[1].is_oa	True
locations[1].source.id	https://openalex.org/S4306400194
locations[1].source.issn
locations[1].source.type	repository
locations[1].source.is_oa	True
locations[1].source.issn_l
locations[1].source.is_core	False
locations[1].source.is_in_doaj	False
locations[1].source.display_name	arXiv (Cornell University)
locations[1].source.host_organization	https://openalex.org/I205783295
locations[1].source.host_organization_name	Cornell University
locations[1].source.host_organization_lineage	https://openalex.org/I205783295
locations[1].license	cc-by
locations[1].pdf_url
locations[1].version
locations[1].raw_type	article
locations[1].license_id	https://openalex.org/licenses/cc-by
locations[1].is_accepted	False
locations[1].is_published
locations[1].raw_source_name
locations[1].landing_page_url	https://doi.org/10.48550/arxiv.2506.04079
indexed_in	arxiv, datacite
authorships[0].author.id	https://openalex.org/A5101708734
authorships[0].author.orcid	https://orcid.org/0000-0002-8038-9073
authorships[0].author.display_name	Pedro Henrique Martins
authorships[0].author_position	first
authorships[0].raw_author_name	Martins, Pedro Henrique
authorships[0].is_corresponding	False
authorships[1].author.id	https://openalex.org/A5103829551
authorships[1].author.orcid
authorships[1].author.display_name	João Alves
authorships[1].author_position	middle
authorships[1].raw_author_name	Alves, João
authorships[1].is_corresponding	False
authorships[2].author.id	https://openalex.org/A5061762931
authorships[2].author.orcid
authorships[2].author.display_name	Patrick Fernandes
authorships[2].author_position	middle
authorships[2].raw_author_name	Fernandes, Patrick
authorships[2].is_corresponding	False
authorships[3].author.id	https://openalex.org/A5039347839
authorships[3].author.orcid	https://orcid.org/0000-0001-8265-1939
authorships[3].author.display_name	Ricardo Rei
authorships[3].author_position	middle
authorships[3].raw_author_name	Rei, Ricardo
authorships[3].is_corresponding	False
authorships[4].author.id	https://openalex.org/A5019543373
authorships[4].author.orcid	https://orcid.org/0000-0001-6384-5332
authorships[4].author.display_name	M. Amin Farajian
authorships[4].author_position	middle
authorships[4].raw_author_name	Farajian, Amin
authorships[4].is_corresponding	False
authorships[5].author.id	https://openalex.org/A5066205759
authorships[5].author.orcid
authorships[5].author.display_name	Mateusz Klimaszewski
authorships[5].author_position	middle
authorships[5].raw_author_name	Klimaszewski, Mateusz
authorships[5].is_corresponding	False
authorships[6].author.id	https://openalex.org/A5007183487
authorships[6].author.orcid
authorships[6].author.display_name	Duarte M. Alves
authorships[6].author_position	middle
authorships[6].raw_author_name	Alves, Duarte M.
authorships[6].is_corresponding	False
authorships[7].author.id	https://openalex.org/A5070177445
authorships[7].author.orcid	https://orcid.org/0000-0003-4900-630X
authorships[7].author.display_name	José P. Pombal
authorships[7].author_position	middle
authorships[7].raw_author_name	Pombal, José
authorships[7].is_corresponding	False
authorships[8].author.id	https://openalex.org/A5064115506
authorships[8].author.orcid
authorships[8].author.display_name	Nicolas Boizard
authorships[8].author_position	middle
authorships[8].raw_author_name	Boizard, Nicolas
authorships[8].is_corresponding	False
authorships[9].author.id	https://openalex.org/A5093123135
authorships[9].author.orcid
authorships[9].author.display_name	Manuel Faysse
authorships[9].author_position	middle
authorships[9].raw_author_name	Faysse, Manuel
authorships[9].is_corresponding	False
authorships[10].author.id	https://openalex.org/A5103969906
authorships[10].author.orcid
authorships[10].author.display_name	Pierre Colombo
authorships[10].author_position	middle
authorships[10].raw_author_name	Colombo, Pierre
authorships[10].is_corresponding	False
authorships[11].author.id	https://openalex.org/A5030615769
authorships[11].author.orcid	https://orcid.org/0000-0002-7972-7442
authorships[11].author.display_name	François Yvon
authorships[11].author_position	middle
authorships[11].raw_author_name	Yvon, François
authorships[11].is_corresponding	False
authorships[12].author.id	https://openalex.org/A5110781707
authorships[12].author.orcid
authorships[12].author.display_name	Barry Haddow
authorships[12].author_position	middle
authorships[12].raw_author_name	Haddow, Barry
authorships[12].is_corresponding	False
authorships[13].author.id	https://openalex.org/A5103075338
authorships[13].author.orcid	https://orcid.org/0000-0001-6344-7633
authorships[13].author.display_name	José G. C. de Souza
authorships[13].author_position	middle
authorships[13].raw_author_name	de Souza, José G. C.
authorships[13].is_corresponding	False
authorships[14].author.id	https://openalex.org/A5038456766
authorships[14].author.orcid	https://orcid.org/0000-0002-9022-3405
authorships[14].author.display_name	Alexandra Birch
authorships[14].author_position	last
authorships[14].raw_author_name	Birch, Alexandra
authorships[14].is_corresponding	False
authorships[15].author.id	https://openalex.org/A5051693368
authorships[15].author.orcid	https://orcid.org/0000-0001-8282-625X
authorships[15].author.display_name	André F. T. Martins
authorships[15].author_position	middle
authorships[15].raw_author_name	Martins, André F. T.
authorships[15].is_corresponding	False
has_content.pdf	True
has_content.grobid_xml	True
is_paratext	False
open_access.is_oa	True
open_access.oa_url	https://arxiv.org/pdf/2506.04079
open_access.oa_status	green
open_access.any_repository_has_fulltext	False
created_date	2025-10-10T00:00:00
display_name	EuroLLM-9B: Technical Report
has_fulltext	False
is_retracted	False
updated_date	2025-11-28T09:52:06.185452
primary_topic
cited_by_count	0
locations_count	2
best_oa_location.id	pmh:oai:arXiv.org:2506.04079
best_oa_location.is_oa	True
best_oa_location.source.id	https://openalex.org/S4306400194
best_oa_location.source.issn
best_oa_location.source.type	repository
best_oa_location.source.is_oa	True
best_oa_location.source.issn_l
best_oa_location.source.is_core	False
best_oa_location.source.is_in_doaj	False
best_oa_location.source.display_name	arXiv (Cornell University)
best_oa_location.source.host_organization	https://openalex.org/I205783295
best_oa_location.source.host_organization_name	Cornell University
best_oa_location.source.host_organization_lineage	https://openalex.org/I205783295
best_oa_location.license	cc-by
best_oa_location.pdf_url	https://arxiv.org/pdf/2506.04079
best_oa_location.version	submittedVersion
best_oa_location.raw_type	text
best_oa_location.license_id	https://openalex.org/licenses/cc-by
best_oa_location.is_accepted	False
best_oa_location.is_published	False
best_oa_location.raw_source_name
best_oa_location.landing_page_url	http://arxiv.org/abs/2506.04079
primary_location.id	pmh:oai:arXiv.org:2506.04079
primary_location.is_oa	True
primary_location.source.id	https://openalex.org/S4306400194
primary_location.source.issn
primary_location.source.type	repository
primary_location.source.is_oa	True
primary_location.source.issn_l
primary_location.source.is_core	False
primary_location.source.is_in_doaj	False
primary_location.source.display_name	arXiv (Cornell University)
primary_location.source.host_organization	https://openalex.org/I205783295
primary_location.source.host_organization_name	Cornell University
primary_location.source.host_organization_lineage	https://openalex.org/I205783295
primary_location.license	cc-by
primary_location.pdf_url	https://arxiv.org/pdf/2506.04079
primary_location.version	submittedVersion
primary_location.raw_type	text
primary_location.license_id	https://openalex.org/licenses/cc-by
primary_location.is_accepted	False
primary_location.is_published	False
primary_location.raw_source_name
primary_location.landing_page_url	http://arxiv.org/abs/2506.04079
publication_date	2025-06-04
publication_year	2025
referenced_works_count	0
abstract_inverted_index.a	4, 49, 90
abstract_inverted_index.11	27
abstract_inverted_index.24	21
abstract_inverted_index.To	127
abstract_inverted_index.We	47, 65
abstract_inverted_index.an	79
abstract_inverted_index.as	83, 85, 118
abstract_inverted_index.by	18
abstract_inverted_index.in	41
abstract_inverted_index.it	117
abstract_inverted_index.of	15, 34, 52, 77, 88, 124, 138
abstract_inverted_index.on	109
abstract_inverted_index.to	11
abstract_inverted_index.we	133
abstract_inverted_index.LLM	123
abstract_inverted_index.all	20, 135
abstract_inverted_index.and	26, 39, 62, 71, 112, 131, 144, 150
abstract_inverted_index.for	94, 100
abstract_inverted_index.its	125
abstract_inverted_index.the	13, 32, 67, 75, 86, 119, 142, 147, 151
abstract_inverted_index.This	0
abstract_inverted_index.base	143
abstract_inverted_index.data	60, 69
abstract_inverted_index.from	9
abstract_inverted_index.open	43, 121, 129
abstract_inverted_index.that	96
abstract_inverted_index.this	139
abstract_inverted_index.well	84
abstract_inverted_index.Union	24
abstract_inverted_index.being	37
abstract_inverted_index.issue	33
abstract_inverted_index.large	5, 44
abstract_inverted_index.major	136
abstract_inverted_index.model	7
abstract_inverted_index.needs	14
abstract_inverted_index.novel	91
abstract_inverted_index.size.	126
abstract_inverted_index.work,	140
abstract_inverted_index.design	87
abstract_inverted_index.report	1
abstract_inverted_index.tasks,	115
abstract_inverted_index.EuroLLM	30
abstract_inverted_index.dataset	93
abstract_inverted_index.design,	57
abstract_inverted_index.filter,	82
abstract_inverted_index.leading	120
abstract_inverted_index.machine	113
abstract_inverted_index.models,	146
abstract_inverted_index.models.	46
abstract_inverted_index.provide	48
abstract_inverted_index.release	134
abstract_inverted_index.results	104
abstract_inverted_index.scratch	10
abstract_inverted_index.support	12, 128
abstract_inverted_index.trained	8
abstract_inverted_index.AI-based	80
abstract_inverted_index.European	16, 23, 35, 101
abstract_inverted_index.citizens	17
abstract_inverted_index.coverage	99
abstract_inverted_index.covering	19
abstract_inverted_index.creation	76
abstract_inverted_index.dataset.	154
abstract_inverted_index.describe	66
abstract_inverted_index.enhances	97
abstract_inverted_index.existing	42
abstract_inverted_index.language	6, 45, 98
abstract_inverted_index.official	22
abstract_inverted_index.overview	51
abstract_inverted_index.presents	2
abstract_inverted_index.research	130
abstract_inverted_index.training	63
abstract_inverted_index.addresses	31
abstract_inverted_index.adoption,	132
abstract_inverted_index.filtering	72
abstract_inverted_index.including	55, 74, 141
abstract_inverted_index.languages	25, 36
abstract_inverted_index.pipeline,	73
abstract_inverted_index.synthetic	92, 152
abstract_inverted_index.tokenizer	56
abstract_inverted_index.EuroFilter	148
abstract_inverted_index.Evaluation	103
abstract_inverted_index.additional	28
abstract_inverted_index.benchmarks	111
abstract_inverted_index.collection	70
abstract_inverted_index.components	137
abstract_inverted_index.filtering,	61
abstract_inverted_index.languages.	29, 102
abstract_inverted_index.EuroFilter,	78
abstract_inverted_index.EuroLLM-9B,	3
abstract_inverted_index.classifier,	149
abstract_inverted_index.competitive	107
abstract_inverted_index.demonstrate	105
abstract_inverted_index.performance	108
abstract_inverted_index.procedures.	64
abstract_inverted_index.translation	114
abstract_inverted_index.underserved	40
abstract_inverted_index.EuroLLM-9B's	53, 106
abstract_inverted_index.development,	54
abstract_inverted_index.establishing	116
abstract_inverted_index.multilingual	81, 110
abstract_inverted_index.pre-training	68
abstract_inverted_index.European-made	122
abstract_inverted_index.architectural	58
abstract_inverted_index.comprehensive	50
abstract_inverted_index.post-training	95, 153
abstract_inverted_index.specifications,	59
abstract_inverted_index.underrepresented	38
abstract_inverted_index.instruction-tuned	145
abstract_inverted_index.EuroBlocks-Synthetic,	89
cited_by_percentile_year
countries_distinct_count	0
institutions_distinct_count	16
citation_normalized_percentile