I’m considering running a DeepSeek model locally rather than sending every prompt to a hosted service. The appeal is more control over private data, but I’m unsure what hardware is actually needed for a usable experience and how much quality I’d give up compared with the online versions.

For anyone who has tried it, what machine are you using, and what do you use the local model for? I’d especially like to hear whether local inference feels practical for everyday questions or mainly makes sense for experiments.