unknown

A developer successfully runs DeepSeek-r1:1.5b on an old laptop with minimal specs using Ollama and Docker. While the setup works for basic tasks and automation workflows, it's slow on complex prompts (up to 10 minutes) but offers cost savings and privacy benefits over paid APIs. The author provides detailed hardware specs, installation steps, and test results showing the model's reasoning capabilities despite performance limitations.

My Laptop Is a Toaster, But It Runs an LLM

Self-hosted refers to the practice of running your own applications and services on your own servers or devices, giving you full control over your data and how it is managed.

selfhosted

I turned a decade-old laptop into a self-hosted LLM server running DeepSeek-r1:1.5b, and it surprisingly holds its own in real-world automations with n8n.

I turned a decade-old laptop into a self-hosted LLM server running DeepSeek-r1:1.5b, and it surprisingly holds its own in real-world automations with n8n.

This is one of the most incapable Ai setup filled with verbose. There is no point in trying this on a local setup. A complete time waste in reading the article, as it lacks how the ai is interacting with n8n and what is the actual use case.

Total waste of time. As a fun experiment, totally cool.

Unfortunately, this post doesn’t actually show a use‑case for DeepSeek and n8n, so like the other comments it’s not very useful.
However, if you read it in a positive light, it demonstrates that DeepSeek can run even on a heavily‑used laptop with older specs, meaning you still have a chance to get something out of it by automating it. I think the idea of reusing old hardware instead of throwing it away is valuable.
I don’t really want to keep using a laptop that’s this worn out, but does this machine actually work?

Sounds interesting. I too have the hardware which is decade old. Shall give it a try. I plan to set it up as a small home-tutor or a general knowledge playground for my two year kid.