Running Text Embedding Models on Jetson Nano for RAG

S_Anv · reddit · 2026-07-30

The poster is asking if anyone has experience running text embedding models (like microsoft/harrier-oss-v1-0.6b) on a Jetson Nano (2GB/4GB). The goal is to build a local RAG (Retrieval-Augmented Generation) system, and they are looking for insights on the actual inference speed and feasibility on this edge hardware.

Original post →

More from Infra

Infra channel →