When done, click the tiny robot icon on the Obsidian sidebar—it opens the chat pane where you can interact with the LLM.

Orin also includes dedicated Deep Learning Accelerators (NVDLA) for steady, power-efficient inference, keeping the GPU free ...