Selfhost your own local LLM with Gemma 4 and llama.cpp
This article will show you how to setup and run your own selfhosted Gemma 4 with llama.cpp – no cloud, no subscriptions, no rate limits.
How to route ComfyUI in a subdirectory behind a nginx reverse proxy
Running ComfyUI within a proxied subdirectory is trickier than it looks. This article covers an nginx reverse proxy configuration that works reliably with ComfyUI, including fixes for workflow saving.
Run WhisperX on NVIDIA Pascal GPUs
Learn how to set up WhisperX on Ubuntu 24.04 with an NVIDIA Pascal GPU, including drivers, Conda environment, and CUDA-compatible PyTorch.