With the rise of visual workflow builders for AI automation, users can now choose between running local/private LLMs (like Ollama) or using cloud-based models (OpenAI, Gemini, etc.). Each approach has trade-offs in privacy, speed, cost, and flexibility.
- What are your experiences using private/local LLMs versus cloud-hosted ones?
- Which do you prefer for building AI-powered workflows, and why?
- Are there specific use cases where one clearly outperforms the other?
- What do you think are the minimum integrations or requirements for an automation AI workflow tool to be truly useful?
Curious to hear the community’s thoughts and recommendations!
[link] [comments]