If you don’t have a local GPU but still want to experiment with LLMs, this project might help. I built a minimal setup to run Ollama models directly on Google Colab with almost zero friction. What this repo does • Installs Ollama inside Colab • R...