Contents tagged with #ollama
blogOllama quietly stopped using my AMD iGPU16-Jul-2026Why Ollama suddenly ran everything on the CPU on my AMD APU box, and the one environment variable that brought the GPU back.
blogGenerating Semantic Note Filenames: Taming Local LLMs with Ollama and Gemma 414-Jul-2026How I integrated LLMs into my PKM converter, refactored API abstractions, debugged network/port quirks, and compared local model performance.