Self-hosting LLMs is way easier than you think

No coding needed - just download a frontend and pull the models directly

Yeah it's slow on consumer hardware but the fact that you can run AI locally now is huge

Open source is eating the AI stack