Our eleventh session of 2025 will be held at the space NOISE. Need guidance in getting around? Have a look here https://bit.ly/3zUDYFy, or ask for it at the Info Desk on the ground floor of the main building.

Many projects that build AI applications powered by LLMs just use the market leader's API and call it a day. However, there has never been more choice in LLMs! As of Summer 2025, the 23 LLMs that share a top-20 position for text processing in the LMArena leaderboard are made by 9 different companies. 18 are proprietary and 5 are Open Source. 16 come from US-based companies, 6 from Chinese ones and just 1 from an EU company. In this talk we'll present an overview of this landscape and give you a few criteria to make the best choice for your project, considering model use-case, model performance, cost, vendor lock-in, geopolitical aspects, fine-tuning and when running your model locally
is an option. In a lighter and more fun moment, SysWhite will show one of the market leader's models and Chris Mair will show one of the alternatives models performing text summarization tasks to compare their performance.

Register now!

18:30 Chris Mair (1006.org), Michele Magri (syswhite.dev): What LLM is right for my job?
19:00 Open discussion