NextSilicon builds a software-defined Intelligent Compute Accelerator (Maverick-2) designed to deliver adaptive, real-time hardware acceleration for HPC, AI, and vector database applications. Maverick-2 creates software-defined processor cores and uses telemetry and patented algorithms to self-optimize performance, power consumption, and utilization as applications run. The company says Maverick-2 can achieve over 4x the performance-per-watt of traditional GPUs and more than 20x that of high-end CPUs, while consuming 50–80% less power than traditional GPUs and cutting operational costs by more than half. Maverick-2 natively supports C/C++, FORTRAN, OpenMP, and Kokkos today, with planned integrations for CUDA, HIP/ROCm, and leading AI frameworks. The accelerator is now shipping to dozens of customers with volume shipments beginning in early 2025 to fulfill a strong backlog, and early-adopter programs are available via partners Penguin Solutions and Dell Technologies. NextSilicon cites customers including Sandia National Laboratories, the U.S. Department of Energy, academic research institutions, and commercial users across finance, energy, manufacturing, and life sciences. Established in 2018, the company has grown to over 300 employees globally and reports $303 million in funding to date.

Total raised

$303M

Funding rounds

1

Latest round

Series C

Latest activity

Oct 2024