Learn how to Run LLMs Locally ~ CUNY
Run Local LLMs on Your Laptop or Phone: A workshop led by CUNY Students
New York has committed over $500M to Empire AI to keep our state at the frontier of artificial intelligence, and CUNY sits inside that consortium alongside Columbia, Cornell, and NYU. But "leading" doesn't happen at press conferences or after press releases. It happens when students at schools like CSI can boot a frontier-grade model on their own laptop and start building.
That's what this workshop is about.
We're Ethan and Hussam, two CUNY students who believe the next wave of AI isn't only happening inside San Francisco offices or data centers in Texas. It's happening on-device. Local LLMs have crossed a real threshold: models that needed a server rack two years ago now fit in a few gigabytes of RAM, run entirely offline, and keep every token on your machine. No API keys, no usage caps, no one logging your prompts.
We'll get hands-on with Google's brand-new Gemma 4, released in April built from the same research as Gemini 3. The smallest variant (E2B) runs on a phone. The larger ones (26B MoE, 31B) are better than frontier models from just a year ago.
You'll leave with:
A working local-LLM setup on your own machine
A clear sense of when local beats cloud (and when it doesn't)
Concrete startup directions this unlocks: privacy-first tools for SMBs, on-device education, offline agents
Research angles worth pursuing as a CUNY student, especially as the Empire AI gains traction.
Who it's for: Any CUNY student curious about AI. No prior AI/ML experience required. Bring a laptop or a relatively modern phone.
The thesis: New York has the talent, the schools, the funding, the companies, and the network. CUNY is the largest urban public university in the country. The students who shape what AI becomes should look like the city they come from. Come build with us.
*THIS IS A STUDENT RUN EVENT SEPARATE FROM CUNY AS AN INSTITUTION.