Lugo Provincial Council runs 17 AI tools on open-source and open-weight models on its own servers (MencIA)
The Provincial Council of Lugo, Spain, operates an AI Laboratory in production with 17 tools (8 applications and 9 assistants) running open-source and open-weight models on its own servers, using Ollama, vLLM, LangChain and Open WebUI, with RAG only on Council documents. The 2026-2030 MencIA strategy aims to extend this to the province's 66 small municipalities. Internal measurements cited in the source: daily email management in the Presidency Office fell from about 60 minutes to 5, and analysis of grant calls from about 30 minutes to 2.
Overview
The Provincial Council of Lugo, Spain, operates an AI Laboratory in production with 17 tools (8 applications and 9 assistants) running open-source and open-weight models on its own servers, using Ollama, vLLM, LangChain and Open WebUI, with RAG only on Council documents. The 2026-2030 MencIA strategy aims to extend this to the province's 66 small municipalities. Internal measurements cited in the source: daily email management in the Presidency Office fell from about 60 minutes to 5, and analysis of grant calls from about 30 minutes to 2.
This entry has 12 published fields tied to exact passages in an immutable source capture.
Inspect the highlighted sourceThe challenge
MencIA addresses the dilemma of 'Digital Feudalism', in which public administrations hand control of their data to proprietary 'black box' technologies. The Council wanted small towns to be able to use secure, publicly controlled AI without depending on large technology vendors.
The solution
The Council runs open-source and open-weight AI models on its own infrastructure, so that sensitive information never leaves its servers. The technical stack combines Ollama and vLLM for inference, Docker for containerisation, LangChain for orchestration and Open WebUI for general chat, with purpose-built interfaces; retrieval-augmented generation works only on the Council's own documents. Each application solves a single, well-defined task and returns a short, structured output, and AI is positioned as an assistant, with a person holding the final decision.
Reported business value
Internal measurements cited by the source show daily email management in the Presidency Office falling from about 60 minutes to 5, extraction of insolvency notices taking seconds rather than 15-20 minutes, and analysis of grant calls falling from about 30 minutes to 2. The extension to municipalities and the 2028 and 2030 targets are plans, not delivered results.
Sources
Open any source and check the claim yourself — that is the point of the register.
This record was researched and written with AI assistance, and its claims were checked against the sources above. (EU AI Act art. 50 transparency notice.)
Other government & public sector entries in the register.
GovTech unlocks data insights to improve nationwide services with Databricks
Singapore's Government Technology Agency (GovTech) migrated from an on-premises dashboarding system to the Databricks Data + AI Platform on AWS with Unity Catalog and Delta Lake, cutting dashboard creation time from 90 to 30 days, democratizing data across 50% of corporate divisions in the first year, and saving 8,000 labor hours annually.
Estonia rolls out Bürokratt, an AI-guided virtual assistant network for public services
Bürokratt is a network of chatbots deployed on Estonian public sector institutions' websites, letting people obtain information from institutions and use public and information services via virtual assistants. It is a state-created, AI-based digital assistant that helps institutions deliver modern, efficient, around-the-clock customer service using large language models.
City of Helsinki pilots watsonx.ai for multilingual customer care chatbot translation
The City of Helsinki collaborated with IBM Client Engineering and IBM Consulting on a Proof of Value testing whether IBM watsonx.ai could translate foreign-language questions into Finnish for its customer care chatbot, which supports Finnish, English, and Swedish. Watsonx.ai scored 0.90 on the COMET machine-translation metric for English-to-Finnish translation (versus 0.91 for the incumbent system) and 0.87 for Swedish-to-Finnish (versus 0.86), performing on par with the city's existing translation solution.
Madrid builds VisitMadridGPT on Azure OpenAI to answer tourists in 95+ languages
The City of Madrid, with partner iUrban, built the VisitMadridGPT virtual assistant on Microsoft Azure OpenAI Service to give tourists accurate, real-time, personalized information in more than 95 languages, drawing from the city's official tourism content. Users average 2.8 questions and spend about 4 minutes 18 seconds per session with a low bounce rate, and the city analyzes assistant conversations to identify popular topics and refine its digital marketing strategy.
Was this helpful?
Your feedback helps us improve our use case database
