Published on 2025-02-02 14:11
How ridiculous is it that I had to build an AI with the help of AI to tell me the best AI to help me with a task? 🤣 Thankfully, it only took 10 minutes.
There are now 7 models to choose from in the paid versions of ChatGPT: GPT-4o, GPT-4o with Scheduled Tasks, o1, o3-mini, o3-mini-high, GPT-4o Mini, GPT-4 (Legacy).
Besides the fact that these names sound like rejected Star Wars droid models (I mean, GPT-4o and o3-mini-high sound like droids that didn’t make the cut), figuring out which one to use for which task is not for mere mortals like many of us.
So I asked ChatGPT to give me the instructions to build a custom GPT (AI built to do a specific task) that will tell us which model to use based on the task that we want to do.
Think of it as a friendly AI assistant that asks what you’re trying to do and points you to the right model.
I cut and pasted the instructions it gave me to build the custom GPT and viola! View the demo below. Feel free to try the AI Model Matchmaker GPT yourself, see the link in the comments.
Here’s my best attempt at decoding these models so far, with some help from ChatGPT:
- GPT-4o – Best for general tasks like content creation, reports, and analysis.
- o1 – Great for deep strategic thinking and complex problem-solving.
- o3 – Good for coding, technical tasks, and programming.
- Mini Models – Faster and more efficient for quick tasks, with a trade-off in depth.
- Mini-high Models – More powerful than Minis, great for complex tasks that still need speed.
Why couldn’t OpenAI simply build in these instructions to choose the best model for us? 🤔 The whole situation is absurd. I mean, their own AI helped me solve their usability problem! 😂
Eventually, user experience will get prioritized. These AIs will talk with each other and do things together so that we don’t have to orchestrate them. We’ll focus on providing human oversight, strategic thinking, and creativity.
In some sense, that’s ok. It gives us some breathing room to understand, test, learn, and adapt.
This is what happens when AI companies are racing to advance tech while still working out how to make it user-friendly. In the meantime, it will be up to us users to share and learn together.
#OpenAI #ChatGPT #Innovation #UserExperience #o3