The Fifth Elephant 2025 Annual Conference (18 & 19 July)

Machine Learning makes one think of large datacenters and high end hardware. But is it possible to run effective ML workloads on your laptop or your homebrew setup? What type of hardware and devices can you put together and wring some ML juice out of it? Join this Birds of Feather session to find out!

Potential topics for this BoF session

1. Feasibility of running ML models (esp LLMs) on high end laptops or workstations: What hardware configurations are needed? How much would it cost to set this up?
2. What kind of applications and use cases can benefit from this? Does it help with your development workflow or privacy concerns?
3. Which models are suitable for running in this manner? What tools exist to run the models this way?
4. Which communities can I join or take help if I’m trying to run models locally?
5. Are things getter better or worse when it comes to running models locally and solving some usecases with it?

Participants

1. Amiruddin Nagri
2. TBD: Confirm more participants