Successfully got my Lobster Robot to run the deepseek large model deployed locally on my 3060 gaming laptop.

My device is the underpowered 3060 laptop.
The model is deepseek-r1-8b.
This is the distilled version of the deepseek large model, fully capable of handling simple text tasks like collecting and processing information from Telegram groups.
It greatly reduces costs and alleviates 50% of token anxiety.