📊 Full opportunity report: AMÁLIA · The Three Hard Questions. on ThorstenMeyerAI.com — validation score, market gap, and execution plan.
TL;DR
Portugal’s state-funded AMÁLIA LLM is now operational, outperforming some benchmarks. However, experts question its openness, native data sufficiency, and optimization goals, highlighting broader issues in European sovereign-LLMs.
Portugal’s €5.5 million investment in the AMÁLIA large language model has resulted in an operational model that outperforms previous benchmarks in European Portuguese tasks, but experts are raising three fundamental questions about its openness, native-language data, and strategic goals.
AMÁLIA was developed by a consortium of approximately 60 researchers from Portugal’s leading institutions, including NOVA and IST, and was announced in December 2024. The model, based on a continuation of the EuroLLM multilingual foundation, was completed in September 2025 and is now accessible to around 450,000 academic users via the FCT’s IAedu platform. It handles text-only input and is knowledge-updated to the end of 2023, with a final version expected in June 2026.
Technically, AMÁLIA is not trained from scratch but builds on an existing multilingual European model, with approximately 5.8 billion tokens from Portuguese sources—mainly the national web archive—and about 17-18% of supervised fine-tuning data being Portuguese. Benchmarks show it surpasses most open models on European Portuguese tasks and beats Qwen 3-8B on many benchmarks, although it still trails on some specific tests like ALBA, the team’s primary benchmark.
AMÁLIA
The three hard
questions.
Portugal spent €5.5M to build a European Portuguese LLM. The base version is operational, the benchmarks beat Qwen 3-8B on most pt-PT tasks. So why are the most important questions still unanswered?
Last month, Duarte O.Carmo published the sharpest public analysis of AMÁLIA — Portugal’s state-funded European Portuguese large language model. He prefaces his critique with the necessary diplomatic apparatus before doing what almost nobody else in the European-sovereign-LLM discourse has been willing to do publicly: asking hard questions about whether the work, as released, actually does what it set out to do. This piece is a structural extension of his analysis. The AMÁLIA case study exposes three hard questions every national LLM effort needs to answer publicly — and the broader European sovereign-LLM movement has been operating without explicit answers to any of them.
Three questions every national LLM effort needs to answer publicly.
Duarte O.Carmo’s framing maps cleanly onto the structural argument. Each question lands specifically in AMÁLIA — and the broader European sovereign-LLM movement has been operating without explicit answers to any of them.
The three questions form a structural feedback loop. Q3 (optimization target) determines Q2 (data volume needed) which conditions Q1 (openness sufficient for community contribution). The European sovereign-LLM movement collectively benefits from these questions becoming standard methodology disclosure, not exceptional critique.
Advanced Language Tool Kit: Teaching the Structure of the English Language
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
107 billion tokens. 5.8 billion clearly pt-PT.
The structurally tractable question with a structurally surprising answer. For a model whose entire stated purpose is European Portuguese prioritization, the native-language share of extended pre-training is 5.5%. The implications cascade into every other question.
MAYAPHILOS 224 Words Brazil Portuguese and English Talking Flash Cards for Toddlers, Autism Sensory Toys, Portuguese Language Learning Educational Montessori Speech Therapy Toys Gifts for Kids
【English and Portuguese Learning】The talking flash cards contain 510 sight words with 31 themes, including letters, numbers, animals,…
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
The Olmo standard. AMÁLIA’s current state.
Allen Institute for AI’s Olmo project defines what “fully open” operationally requires. Olmo doesn’t lead frontier benchmarks. That’s not the point. The point is to be the structural reference for openness. AMÁLIA’s “fully open source” claim should track to the operational standard.
AI model fine-tuning tools
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Four strategic positions. AMÁLIA between two and three.
Approximately €100M+ in publicly disclosed European sovereign-LLM funding across the major initiatives. The structural question every project faces: what is the actual competitive position you’re staking? Four options — none mutually exclusive — but each requiring different commitments.
European language NLP software
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Three standards. For AMÁLIA and the movement.
The structural critique generalizes beyond AMÁLIA. Italy, France, Germany, Switzerland, the OpenEuroLLM consortium, and every subsequent national project benefit from public discourse holding national LLM efforts to operational standards on openness, data accounting, and strategic positioning.
The European sovereign-AI agenda is a serious strategic project that deserves serious public discourse. O.Carmo’s analysis is what serious public discourse looks like. Appropriately diplomatic. Structurally rigorous. Willing to ask the hard questions in public when the public investment justifies it. More of this is needed — across every European sovereign-LLM project, not just AMÁLIA.
Broader Implications for European Sovereign-LLMs
This development highlights the challenges faced by European countries in developing independent language models. The questions about openness, native data sufficiency, and strategic goals are central to understanding whether these models can truly serve national interests or are limited by technical and strategic constraints. Portugal’s experience exemplifies the broader structural issues across Europe, where investments are substantial but clarity on purpose and openness remains elusive, impacting policy and research credibility.
European Sovereign-LLM Initiatives Face Common Challenges
Across Europe, countries like Italy, Germany, France, and Norway are investing heavily in developing their own large language models, often with public funds. Most are operating with similar technical strategies—either training from scratch or building on multilingual foundations—and face questions about how open these models truly are, how much native-language data is enough, and what their primary objectives should be. Portugal’s AMÁLIA is a prominent case because of its public funding and national scope, making these questions particularly salient at the policy level. The discourse so far has focused on individual model performance, but experts argue that the structural patterns and strategic choices are equally critical to assess.
“AMÁLIA demonstrates impressive performance, but it also exposes fundamental questions about what we really want from these models.”
— Duarte O.Carmo, AI researcher
Unanswered Questions About AMÁLIA’s Openness and Goals
It remains unclear how open the final version of AMÁLIA will be, especially regarding access, licensing, and transparency. Additionally, the strategic objectives—whether the model aims primarily for academic, commercial, or governmental use—are still under discussion. The extent to which native-language data suffices for future improvements is also uncertain, as the current Portuguese data sources are limited relative to the model’s ambitions.
Next Steps for Portugal’s AMÁLIA Development and Evaluation
The final version of AMÁLIA is scheduled for release in June 2026, after which detailed evaluations and transparency reports are expected. Researchers and policymakers will scrutinize its openness, native-language data integration, and strategic purpose, influencing future investments and European sovereign-LLM policies. Ongoing critique and analysis from the academic community will shape the discourse in the coming months.
Key Questions
What are the main technical features of AMÁLIA?
AMÁLIA is based on a continuation of the EuroLLM multilingual foundation, trained on approximately 107 billion tokens, with 5.8 billion tokens from Portuguese sources. It is designed for text-only tasks and is optimized for European Portuguese benchmarks.
Why are questions about openness important for AMÁLIA?
Openness determines how accessible and transparent the model is, impacting trust, collaboration, and strategic sovereignty. It also affects whether the model can be widely adopted or remains restricted.
How does AMÁLIA compare to other European models like Italy’s Minerva?
Unlike Minerva, which was trained from scratch on Italian and English data, AMÁLIA builds on an existing multilingual foundation, with a focus on Portuguese. Benchmarks show it outperforms some models but still has gaps, especially on specific tasks like ALBA.
What are the broader implications for European AI policy?
The case of AMÁLIA illustrates the need for clear strategic objectives, transparency, and sufficient native data in national LLM initiatives, which will influence future funding, regulation, and development across Europe.
Source: ThorstenMeyerAI.com
