Model self-hosting

| WIP: coming soon

Meanwhile, visit our managed hosting platform CoGenAI for shared and dedicated model hosting and inference, the easiest way to deploy and use LLMs.