Self-Hosting

Let’s compare prices for top-level consumer GPUs, that are suitable for LLMs in particular and AI in general. Specifically I’m looking at RTX-5080 and RTX-5090 prices. They have slightly dropped.

Ollama’s GPT-OSS models have recurring issues handling structured output, especially when used with frameworks like LangChain, OpenAI SDK, vllm, and others.

Large Language Models (LLMs) are powerful, but in production we rarely want free-form paragraphs. Instead, we want predictable data: attributes, facts, or structured objects you can feed into an app. That’s LLM Structured Output.

For KDE Plasma fans, two Linux distributions frequently come up in discussion: Kubuntu and KDE Neon. They may appear similar - both ship with KDE Plasma as the default desktop, both are based on Ubuntu, and both are friendly to newcomers.

Here I am comparing how much VRAM new version of Ollama allocating for the model vs previous Ollama version. The new version is worse.

This guide will walk you through the process of changing the static IP address on an Ubuntu Server.

Ollama has quickly become one of the most popular tools for running LLMs locally. Its simple CLI, and streamlined model management have made it a go-to option for developers who want to work with AI models outside the cloud. But as with many promising platforms, there are already signs of Enshittification:

Mumble is a free and open-source voice over IP (VoIP) application designed primarily for real-time voice communication. It uses a client-server architecture where users connect to a shared server to talk with each other.

Locally hosted Ollama allows to run large language models on your own machine, but using it via command-line isn’t user-friendly. Here are several open-source projects provide ChatGPT-style interfaces that connect to a local Ollama.

GPU load monitoring applications: nvidia-smi vs nvtop vs nvitop vs KDE plasma systemmonitor.

Here’s a step-by-step walkthrough of installation of a 3-node K3s cluster on bare-metal servers (1 master + 2 workers).

Comparing self-hosting Kubernetes distributions for hosting on bare-metal or home servers, focusing on ease of installation, performance, system requirements, and feature sets.

I’m comparing self-hosted Kubernetes variants that suit the Ubuntu-based homelab with 3 nodes (16GB RAM, 4 cores each), focusing on ease of setup and maintenance, support for persistent volumes and LoadBalancers.

GitHub Actions is an automation and CI/CD platform within GitHub, used to build, test, and deploy your code based on events like pushes, pull requests, or on a schedule.

Here’s a Docker Compose cheatsheet with annotated examples to help you master Compose files and commands quickly.