Hands-On: Local LLMs for study and research
Content | In today’s AI-driven academic world, protecting sensitive research data, from confidential drafts to unpublished results, is crucial for trust and compliance. By using open large language models (LLMs), you take control of your data, reduce dependencies on external providers, and implement ethically responsible AI practices that meet your university’s data protection standards. This workshop offers students, PhD students, researchers, and all university members the opportunity to use LLMs without compromising data security. Unlike commercial providers that process data externally, open, local LLMs run entirely in secure environments or even on your laptop. Your research, experiments, and communication remain under your control at all times (the limitations of open models lie less in the performance of the models than in the available hardware). This is not just a technical advantage, but a crucial step towards ethical, GDPR-compliant research that protects institutional trust and scientific integrity. In this hands-on workshop, you will learn to select the right LLM for your research goals by weighing performance, size, and data protection, install and configure corresponding applications on your laptop, and use them effectively for, e.g., text assistance, code creation, and data analysis. The goal is to set up a functional, AI assistant tailored to your academic work. No prior knowledge is required, but the workshop Artificial Intelligence for Scientific Work: Ethics, Responsibility, and Practical Application is recommended. A GWDG LLM API Key (“API access to our chat service”) is beneficial and can be booked free of charge for members of the University of Leipzig at https://kisski.gwdg.de/en/leistungen/2-02-llm-service/. |
Target Group | Students of advanced semesters, doctoral students, scientific staff from all disciplines |
| Duration | 90 minutes |
| Prerequisite | none, please bring your own device |
| Date |
|
| Place | Bibliothek Medizin/Naturwissenschaften, Liebigstr. 23/25, Raum 102 |
| Language | English |
| Registration | |
| Kontakt | M. Czygan |