Go EuropeAIdirectory

Evidence-backed European AI register

AMALIA — Automatic Multimodal Language Assistant with Artificial Intelligence

Reviewed identity, roles, model listings, pricing and evidence for AMALIA — Automatic Multimodal Language Assistant with Artificial Intelligence.

Dataset compiled 2026-08-27 · evidence over inference

national initiativeconsortiummodel lab3 model listings0 token-price rows

Reviewed identity

Headquarters scope
EU headquarters
Headquarters record
Lisbon (coordinated by NOVA University Lisbon)
Legal entity
university consortium under a national programme
Ownership
public research
Founded
2024
Confidence
high
Last verified
2026-08-22

Links

Official website

Open interactive profile

Evidence boundaries

Organisation listings identify a reviewed canonical record. Model listings show an association in the source register; they do not by themselves establish model creation, current API availability, hosting location or infrastructure ownership.

Model listings

  1. AMALIA-9B-0626-DPOcatalogued with this organisation · origin not asserted
    Type
    text
    Parameters
    9B
    Context
    32,000 tokens
    Languages
    European Portuguese (pt-PT), English
    Licence
    Apache-2.0
    Release
    2026-07-01
  2. AMALIA-9B-0626-SFTcatalogued with this organisation · origin not asserted
    Type
    text
    Parameters
    9B
    Context
    32,000 tokens
    Languages
    European Portuguese, English
    Licence
    Apache-2.0
    Release
    2026-07-01
  3. AMALIA-VL-DPOcatalogued with this organisation · origin not asserted
    Type
    multimodal
    Parameters
    unknown
    Context
    unknown
    Languages
    European Portuguese, English
    Licence
    unknown
    Release
    2026

Profile note

Portugal's national LLM, launched 1 July 2026 — the first open large model built specifically for European Portuguese, and a striking cost contrast to larger national programmes (roughly €7M). Deliberately NOT a consumer chatbot. Consortium: NOVA University Lisbon (coordinator), Instituto Superior Técnico, and the universities of Coimbra, Porto and Minho, with Beira Interior, Évora and ISEL contributing. Community quantisations already exist (layerx-labs GGUF, teex-pt MLX 4-bit).

Record sources

  1. https://huggingface.co/amalia-llm/AMALIA-9B-0626-DPO
  2. https://huggingface.co/amalia-llm/AMALIA-9B-0626-SFT
  3. https://github.com/AMALIA-LLM/AMALIA
  4. https://portugal.gov.pt/en/gc25/communication/news/llm-amalia-shows-portugals-potential
  5. https://www.it.pt/News/NewsPost/5258
  6. https://www.fct.unl.pt/en/news/2026/07/amalia-portuguese-llm-contribution-nova-fct-now-available
  7. https://oecd.ai/en/dashboards/policy-initiatives/amalia-automatic-multimodal-language-assistant-with-artifici…
  8. https://www.it.pt/