AI_dev: Open Source GenAI & ML Summit Europe 2025

In the rapidly evolving landscape of artificial intelligence, Vision-Language Models (VLMs) have emerged as powerful tools capable of understanding and interpreting both visual imagery and natural language. In this talk, we’ll dive into VLMs and how they work without getting bogged down in tech jargon. 

In October 2024 the Open Source Initiative (OSI) unveiled the Open Source AI Definition (OSAID) v.1, concluding a multi-year, world-wide community effort.

Recent research has raised concerns about LLM-generated code exhibiting significant similarity to their training data, raising potential legal issues with incompatible software licenses. While Xu et al. established a benchmark for evaluating this phenomenon through their LiCoEVAL benchmark, showing small but significant portions of LLM outputs containing "notably similar" code to existing open-source implementations, these findings were limited by the scope of their reference dataset.

Enabling LLMs to improve their outputs by using more test-time computation is a critical step towards building generally self-improving agents that can operate on open-ended natural language. The scaling of inference-time computation in LLMs, with a focus on answering the question: if an LLM is allowed to use a fixed but non-trivial amount of inference-time compute, how much can it improve its performance on a challenging prompt? Answering this question has implications not only on the achievable performance of LLMs, but also on the future of LLM pretraining and how one should tradeoff inference-time and pre-training compute. Despite its importance, little research attempted to understand the scaling behaviors of various test-time inference methods. 

In this 1-hour workshop, participants will have the opportunity to learn how to achieve robust observability for Large Language Models (LLMs) while safeguarding sensitive data. As LLMs become integral to production systems, monitoring their performance, usage, and costs is essential, and so is protecting user privacy! This session addresses these challenges using open-source tools such as OpenTelemetry + OpenLIT, and Prometheus + Grafana.

With the rise of on-device intelligence, the push to run LLMs on edge hardware — phones, Raspberry Pis, even microcontrollers — is accelerating. At the heart of this revolution is quantization: the art of shrinking models without shrinking their intelligence.