Lamb Labs
Lightning fast chips with hardcoded AI models
About
Lamb Labs is building MPUs (Model Processing Units), custom chips that hardcode LLM weights directly into the chip. Unlike GPUs, which move weights between memory and compute, MPUs keep the model on-chip, targeting up to 20,000+ tok/s and a 63× higher intelligence per watt.
Build your own investing agent
GPAgent keeps YC listings public and neutral. Fund-specific scoring, notes, and workflow state live in each customer workspace.
Join the GPAgent waitlistDetails
Status
Active
Stage
Early
Regions
United Kingdom, Europe