📊 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 large language model is now operational, outperforming several benchmarks. However, fundamental questions about openness, native-language data sufficiency, and optimization goals remain unresolved, impacting future European LLM strategies.
Portugal’s €5.5 million investment in the AMÁLIA large language model has resulted in an operational system that surpasses some benchmarks for European Portuguese tasks, but key questions about its openness, data sufficiency, and objectives remain unresolved, raising concerns about the country’s AI strategy and broader European efforts.
AMÁLIA, developed through a consortium involving about 60 researchers across Portugal’s top research institutions, is a continuation of a multilingual European foundation model, not trained from scratch. It was officially completed in September 2025, with the final version expected by June 2026. The model performs well on European Portuguese benchmarks, outpacing previous open models and beating Qwen 3-8B on most Portuguese tasks, though it still trails on some key benchmarks like ALBA.
Despite its technical achievements, the project faces scrutiny over three core questions: How open is ‘fully open’ in practice? How much native-language data is enough for effective modeling? And what should the model’s primary optimization goals be? These questions are central to evaluating the strategic value of AMÁLIA and similar European projects, yet they remain largely unaddressed publicly, according to recent analyses.
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.
European Portuguese NLP training data
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 openness assessment 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.
Evals for AI Engineers: Systematically Measuring and Improving AI Applications
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.
Implications for European Sovereign-Language AI Strategies
The questions surrounding AMÁLIA reflect broader challenges faced by European countries in developing autonomous, open, and effective language models. The answers will influence policy decisions, funding allocations, and the future of AI sovereignty across the continent. The ongoing debates highlight the need for transparent criteria on openness, native-language data use, and strategic objectives, which are critical for ensuring these models serve national interests and uphold European values.
European Sovereign-Language Models and Strategic Challenges
Several European nations have launched or announced large language model initiatives, including Italy’s Minerva, Germany’s Aleph Alpha, France’s Mistral, and others across Scandinavia. These projects share a common structural challenge: balancing openness with proprietary considerations, determining native-language data sufficiency, and defining clear goals for model deployment. The case of AMÁLIA exemplifies these issues, as Portugal’s investment aims to bolster national AI capacity but faces scrutiny over strategic transparency and technical choices. The discourse remains focused on individual models rather than the overarching structural patterns shaping European AI sovereignty.
“AMÁLIA is an impressive piece of work. But the important questions about its openness, native data, and objectives remain unanswered.”
— Duarte O.Carmo
Unresolved Strategic and Technical Questions
It remains unclear how open AMÁLIA truly is in practice, given the proprietary elements of its development process. The sufficiency of native Portuguese data for future model improvements is also debated, with some experts questioning whether current data volumes are adequate for robust language understanding. Additionally, the primary objectives—whether to maximize openness, performance, or strategic sovereignty—are still under discussion within the Portuguese research community and policymakers.
Upcoming Milestones and Ongoing Evaluations
The final version of AMÁLIA is expected by June 2026, which will provide further insights into its capabilities and strategic positioning. Over the next 12-24 months, researchers and policymakers will scrutinize its performance, openness, and data use, shaping future investments and policy frameworks. Public debates and transparency initiatives are likely to increase as the project matures, influencing broader European sovereign AI strategies.
Key Questions
What makes AMÁLIA different from other European language models?
AMÁLIA is a continuation of a multilingual European foundation model, not trained from scratch, and is publicly funded by the Portuguese government. It outperforms previous open models on Portuguese benchmarks but faces questions about its openness and strategic objectives.
Why are the three questions about openness, data, and goals important?
These questions determine how European models can be developed responsibly, transparently, and in line with national interests. They impact policy, strategic sovereignty, and the technical effectiveness of future models.
What are the risks of not addressing these questions publicly?
Without transparent answers, there is a risk of misaligned expectations, inefficient use of public funds, and reduced strategic control over AI developments that could influence national and European interests.
Will the final version of AMÁLIA address these questions?
It is uncertain. While the final version is expected in June 2026, whether it will clarify these strategic issues remains to be seen, as ongoing debates continue within the research and policy communities.
How does AMÁLIA compare to other European models like Minerva or Mistral?
Unlike Minerva, which was trained from scratch, AMÁLIA builds on a multilingual foundation. Its performance is competitive on Portuguese benchmarks, but its strategic approach and openness are still under evaluation, similar to other European projects.
Source: ThorstenMeyerAI.com
