📊 Full opportunity report: ALIA. The Spanish answer. on ThorstenMeyerAI.com — validation score, market gap, and execution plan.
TL;DR
Spain’s ALIA project has released a 40-billion-parameter multilingual AI model, marking Europe’s largest publicly funded national AI effort. While operationally credible, benchmark results suggest a structural capability gap compared to Llama 2, raising questions about strategic positioning.
Spain has officially launched ALIA, its largest publicly funded AI initiative, releasing a 40-billion-parameter multilingual model trained on over 9.37 trillion tokens across 35 European languages and 92 programming languages. Learn more about the strategic implications of hyperscaler investments. The project, led by the Barcelona Supercomputing Center and coordinated by Spain’s Secretary of State for Digitalisation and Artificial Intelligence, aims to establish Spain as a strategic player in European AI development, emphasizing multilingual and Spanish-language capabilities.
Funded with over €240 million in public investment, ALIA (Artificial Linguistic Intelligence for Administration) is designed to support Spanish government initiatives and promote AI adoption across Spain and the broader Spanish-speaking world. The model was trained on MareNostrum 5’s high-performance GPU infrastructure, utilizing 4,480 NVIDIA H100 GPUs. It was released under the Apache License 2.0 on HuggingFace on April 22, 2025, with the goal of fostering transparency and open collaboration.
Benchmark results indicate that ALIA-40B underperforms compared to Llama 2, with accuracy scores of approximately 51.77% on XNLI in English versus Llama 2’s 66%, and 81.53% on SQuAD in English versus Llama 2’s 93-94%. These empirical results confirm a structural capability gap, aligning with prior analysis suggesting that the project’s current scale and funding produce sub-Llama-2 performance levels. The project positions itself as a Position 3 strategic effort, focusing on Spanish-language dominance and multilingual coverage, rather than aiming to outperform global models like Llama 2.
ALIA.
The Spanish
answer.
€240M+ Spanish public funding · ALIA-40B + Salamandra family · 9.37T tokens · 35 European languages + 92 programming languages · MareNostrum 5 · Apache 2.0 release. The largest publicly funded European national-AI project by cumulative scope — and the empirical test case for the Position 1 vs Position 3 strategic-positioning argument.
This is the tenth standalone essay in the European sovereign-LLM track and the third Tier 2 expansion piece. ALIA is Spain’s institutional answer — the largest EU member state by GDP not yet documented in the track. The project markets itself as Position 1 + Position 2 simultaneously — “Europe’s first public multilingual foundational model.” The benchmark evidence (ALIA-40B 51.77% XNLI_en vs Llama 2 66%) confirms the structural capability gap from Finding 1 of the synthesis essay. The Position 3 framing — Martorell’s “most widely adopted in the Spanish-speaking world” — is operationally honest. €90M MareNostrum 5 upgrade + €150M company integration = €240M+ cumulative scope. Apache 2.0 open-source release + AESIA validation + co-official languages oversampling. Both can be true at once. The Spanish public discourse would benefit from explicit Position 3 strategic positioning.
Six models. Apache 2.0.
The ALIA family operates as a tiered model portfolio. ALIA-40B is the flagship at 40 billion parameters; the Salamandra family scales down to 7B, 2B and instruct-tuned variants; mRoBERTa provides the foundational multilingual baseline. All released under Apache License 2.0 on April 22, 2025 at the HispanIA 2040 event — “Public Code, Public Money” approach.
multilingual
MN5 LLM
edge
target
instruct
encoder
Natural Language Processing with Transformers, Revised Edition
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Four official. Oversampled by factor of 2.
ALIA’s distinctive multilingual coverage strategy. The four co-official Spanish languages are oversampled by factor of 2 in the training corpus — structurally distinct from Apertus’s broad 1,811-language coverage approach. The strategy targets deep coverage of Spanish co-official languages rather than maximum language breadth.
Mini AI Voice chatbot, smart Voice Assistant, Multiple AI Models, Emotional Interaction, 100+ Stickers, Suitable for Home and Office use, (Black)
1. Emotional Interaction: This chatbot can recognise and respond to your emotions, offering a more personalised and human-like…
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
ALIA-40B vs Llama 2. 14-point gap.
The empirical evidence Finding 1 of the synthesis essay needed. ALIA-40B at 40 billion parameters with €240M+ public funding and 8+ months MareNostrum 5 training achieves performance below Llama 2 — a 2023 frontier model released approximately 18 months before ALIA-40B. The capability gap is real and consistent with six of seven prior national-project answers documented in the track.
Cursor AI for Programmers: Build, Debug, Refactor, and Ship Code Faster with AI Without Losing Control
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Two pilots. Public administration deployment.
The operational deployment targets that validate the Position 3 + Position 4 framing. Public administration deployment is the structurally credible Position 3 + Position 4 strategic positioning — captive demand from Spanish public institutions where Spanish-language specialization is operationally distinctive.
The work is real across the Spanish ALIA case. €240M+ public funding committed. 40B parameter from-scratch model trained on 9.37 trillion tokens. Salamandra family released under Apache 2.0. AESIA validation aligned with EU AI Act transparency standards. Two pilot applications shipped — Tax Agency chatbot and primary care medicine heart failure diagnosis. The Position 1 framing is operationally misleading. ALIA-40B performance below Llama 2 confirms the structural capability gap. The Position 3 framing is operationally honest — Spanish-speaking world adoption, co-official languages oversampling, public administration deployment. Both can be true at once. The Spanish public discourse would benefit from explicit Position 3 strategic positioning.
AI Systems Performance Engineering: Optimizing Model Training and Inference Workloads with GPUs, CUDA, and PyTorch
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Implications of ALIA’s Strategic Positioning and Performance
While ALIA represents Europe’s largest publicly funded national AI project in terms of scale and scope, its benchmark performance indicates a structural gap compared to leading models like Llama 2. The project’s emphasis on Spanish-language and multilingual coverage aligns with Spain’s strategic goal to foster widespread adoption within the Spanish-speaking world, rather than competing for top performance globally. This approach underscores a shift towards operational relevance and regional influence over raw benchmark metrics, which could influence European AI policy and national sovereignty considerations.
Background on European Sovereign AI Strategies and ALIA’s Role
Spain’s ALIA project is part of a broader European effort to develop sovereign AI capabilities, with previous initiatives including Portugal’s AMÁLIA, Italy’s Minerva, and pan-European projects like OpenEuroLLM and Mistral. Unlike these efforts, ALIA is the largest in scale within Spain, reflecting a strategic decision to prioritize multilingual and regional relevance. The project is also a response to the European Union’s push for independent AI development, with €90 million allocated for infrastructure upgrades and €150 million dedicated to integrating ALIA into industry applications.
Prior to ALIA, Spain had invested in smaller language-specific projects such as AINA (Catalan) and ILENIA (Spanish and co-official languages), but the current effort marks a significant escalation in ambition. The project’s leadership emphasizes that the goal is to maximize adoption in the Spanish-speaking world, rather than achieving the highest benchmark scores globally, aligning with the Position 3 strategic profile.
“The goal is not to be the best-performing LLM in the world, but the most widely adopted in the Spanish-speaking world.”
— Josep M. Martorell, ALIA project lead
Operational Performance vs. Strategic Goals in Focus
While benchmark results confirm a performance gap relative to Llama 2, it remains unclear how ALIA’s operational utility and adoption will compare in real-world applications. For insights on industry trends, see the latest analysis on hyperscaler Capex. The extent to which ALIA will succeed in fostering widespread Spanish-language AI adoption, and how it will evolve to improve performance, are still developing. Additionally, the strategic implications of positioning as a Position 3 project versus a global competitor are subject to ongoing debate within European AI policy circles.
Next Steps for ALIA’s Development and Adoption
Further benchmarking and real-world testing of ALIA are expected to clarify its operational capabilities. The project team plans to continue refining the model, expanding multilingual and domain-specific datasets, and promoting industry integration within Spain. Policymakers and stakeholders will monitor how ALIA’s adoption progresses across government agencies and private sector applications, alongside developments in European AI policy and funding allocations.
Key Questions
What is the main purpose of Spain’s ALIA project?
ALIA aims to develop a multilingual AI model focused on Spanish-language and regional European coverage, prioritizing widespread adoption over benchmark performance.
How does ALIA compare to other European AI initiatives?
While it is the largest publicly funded European national AI project in scale, benchmark results suggest it currently lags behind models like Llama 2 in raw performance, reflecting a strategic focus on regional relevance.
What are the key technical features of ALIA?
ALIA is trained on over 9.37 trillion tokens across 35 European languages and 92 programming languages, using MareNostrum 5’s GPU infrastructure, and was released under an open-source license.
What are the main challenges facing ALIA?
Benchmark results indicate a performance gap compared to leading models, and its success depends on real-world adoption and further model improvements tailored to Spanish and regional needs.
What does ALIA’s development mean for European AI sovereignty?
It demonstrates a significant investment in regional AI capabilities, emphasizing multilingual and Spanish-language coverage, aligning with EU strategies for independent AI development.
Source: ThorstenMeyerAI.com
