Just FYI When building AI agents, at some point, you will need to reduce the context window to minimize costs, increase speed response, and avoid overwhelming the device. Unless you have unlimited amount of tokes, or your device can run a powerful open-source AI model (a.k.a LLM) , in which case you can go bananas, jaja. Any thoughts?

Does anyone have an idea of which Ollama model works best with openclaw and N8N, or it entirely depends on use case ?

Does anyone know how safe it is to start using ecash as a principal method for transferring value?