Skip to content

Details

New location at Brazos Place @ 800 Brazos st. us the entrance on brazos street not 8th.
This is a Hybrid event both in-person and virtual. The in-person event is hosted at Antler downtown Austin.

Address:
Brazos Place @ 800 Brazos st,
The entrance is off of Brazos st.

Talk title: Manu Nicholas - What Your GGUF File Name Does Not Tell You

Talk description: A GGUF file name promises a quantization format, but what your machine actually pays for is bytes. Every generated token reads every weight, so decode speed is close to memory bandwidth divided by model bytes: on an Intel laptop that rule fits at R^2 0.996. This talk opens real model files with llama-roofline, an MIT-licensed Python tool on PyPI, to show what is inside them, then measures what they cost on a laptop. Along the way: why using all 20 threads made generation a quarter slower than 8 while prompt processing got 4.9x faster, and how background load cut measured bandwidth from 50.7 to 14.1 GB/s. Bring a laptop if you want to run it yourself.

Bio: Manu Nicholas Jacob is a hardware engineer at Dell in Austin, where he works in root-cause engineering on enterprise AI servers, looking at why systems fail and what the data actually says about it. He studied electrical and computer engineering at UMass Amherst.
Outside work he runs reproducible measurement studies on edge AI hardware, mostly Raspberry Pi 5 boards and laptops, on what really governs latency and energy: memory bandwidth, thermal limits, and quantization format. He maintains two open-source Python packages, llama-roofline and ml-systems-lab, reviews for the Journal of Open Source Software, and served on the ACM SIGOPS ATC 2026 artifact evaluation committee.

Antler
Antler works with founders through our signature founder residency, adding value immediately, and taking no equity up front. We receive over 160,000 applications each year to build companies with us, and have grown to 18 funds with residencies in 30 cities around the globe. Founders also have the opportunity to pitch our Investment Committee to receive an up to $500K investment, allowing them to build for 12-18 months rather than attempting to raise before they have real traction.

You may also like