Do we actually need 5 AI models, or is one good model enough?
retained text
I've been thinking about this lately. There are so many AI tools now — ChatGPT, Claude, Gemini and others. Some people compare answers from different models, while others stick with one and use it for almost everything. I can see both sides. Using one model is simpler, but getting a second opinion can be useful when the question really matters. Personally, I think it depends on what you're using AI for. But I'm curious what other people think. Do you usually stick with one AI, or do you switch between different models? And has comparing answers ever changed your mind about something?
document fetched 2026-10-12 00:34 · canonical https://reddit.com/r/PromptEngineering/comments/1x3aq9j/do_we_actually_need_5_ai_models_or_is_one_good