# we care about the technical stuff — local inference, quantization, throughput, and the silicon underneath it.
# we’re not here to get rich. but electricity isn’t free and the next GPU won’t buy itself — so someday we’ll sell a thing or two to keep the lab powered and the hardware coming.
# until then it’s a discord, and it’s free. come tinker.
A few days of local inference in one room — hardware, quantization, throughput, and the people who actually run this stuff. We’re working out the city and dates now.
# pre-registration opens 10/1. subscribe to the digest and we’ll send the link the morning it does.
Benchmarks people posted, configs that worked, hardware that didn’t. Summarized and sent once a week — so you can skip the scrollback and still know what happened.