Ever feel like you’re not getting the best results from ChatGPT? It might be because you’re using the wrong model for the job. I just stumbled upon this super useful cheat sheet from a LinkedIn creator that clears it all up.
The original poster points out that knowing which model to use is step one, and I think this breakdown is a game-changer!
Here’s the expert’s guide to picking the right model:
📌 For Quick Answers: GPT-4o
This AI professional says to treat it like a Google search. It’s best for fast answers that don’t need deep reasoning.
- No need for a lot of context.
- Just ask your question directly.
- Example prompt:
“Who’s the CEO of Astronomer?”
🧠 For Deep Reasoning: o3
The author calls this the smartest and best model for reasoning, but it takes a bit longer. The key is to prompt it like you’re briefing an intern.
- Give it as much context as possible.
- Start with the role, then the task, then context and an example.
- The creator shared this killer prompt structure:
“Act like a [expertise]. I want you to [ask], as [format]. Here’s an example of past [ask] I’ve done perfectly: [past example, this is your context].”
🚀 For In-Depth Analysis: Deep Research
This is the one for incredibly detailed results. You start broad, and it helps you narrow things down.
- The author explains that it will ask you follow-up questions.
- You can connect sources like PDFs and URLs.
- Example prompt:
“Pros and Cons of having AI replace everyone’s job. I only want the AI scientists/researchers’ answers.”
Knowing which tool to use is half the battle, and this guide is awesome. The original poster also wrote a full blog post on how to get your ChatGPT set up perfectly.
For all the details and the link to that blog, be sure to check out the full, original post!