chiprook
← Hardware
HardwareOctober 14, 2026, 10:21

Geniatech launches RK1828 and Hailo-10 M.2 AI accelerators for local LLM inference

Geniatech has listed two M.2 2280 AI accelerator cards for edge inference: the AIM-M-R28 based on Rockchip's RK1828 with 20 TOPS and 5GB of in-package DRAM, and the AIM-M-H10 using Hailo-10 with up to 40 TOPS and up to 16GB LPDDR4/4X. Both connect over PCIe and target local LLM, computer vision and audio workloads on embedded systems.

Geniatech launches RK1828 and Hailo-10 M.2 AI accelerators for local LLM inference
#Geniatech#Rockchip#Hailo
Read next
Hardware

MLPerf Inference v6.1: 5.7x per-accelerator gains, 512-GPU run, Vera Rubin results

Hardware

Fujifilm starts selling 40TB LTO-10 tapes for $494

Hardware

Study of 443,000 HDDs over 13 years: drives now last 5-10 years

Hardware

KingKong Technology Open-Sources Jumper, a 22-DoF Crab Robot That Jumps Over 400 mm