Gimlet Labs Raises $80M to Solve AI Inference Bottleneck With Multi-Silicon Cloud

Gimlet Labs closed an $80 million Series A led by Menlo Ventures for its 'multi-silicon inference cloud' — software that distributes AI workloads across diverse hardware (CPUs, GPUs, high-memory systems) simultaneously, claiming 3x-10x inference speedup at equivalent cost and power. Founded by Stanford adjunct professor Zain Asgar, the company launched in October with eight-figure revenue and has since doubled its customer base, including a major model maker and a large cloud provider. Partnerships span NVIDIA, AMD, Intel, ARM, Cerebras, and d-Matrix. The approach addresses a critical industry bottleneck as agentic AI workloads demand increasingly heterogeneous compute.