If you use any of the popular AI chatbot tools like ChatGPT, Gemini, Claude, Copilot, Perplexity, etc., you’ve probably come across a wide variety of models to choose from.
The companies don’t do a great job of explaining why you would use one model instead of another. Most of the explanations sound the same (“This model is great at stuff and you should use it”).
The model that you choose has a huge impact on the quality of your output.
Here is a high-level overview of the common types of models and when you would use them -
- Reasoning Models
- Use for tasks that require logic, mathematical computations, coding, structured decision making, and/or have a lot of steps.
- Deep Research Models
- Use for tasks that require gathering data from a lot of sources and putting it together.
- Image Models
- Use to generate and edit images.
- Audio Models
- Use to generate and edit audio.
- Video Models
- Use to generate and edit video.
- Specialized Models
- This is kind of a catch-all category, but you will come across models that are designed to be used for a specific domain, such as medical, legal, financial, etc.
- General-Purpose Models
- Use for most tasks. They are designed to do well at a broad range of tasks like conversation, simple Q&A, content generation, summarization, language translation, etc.
- If you don’t know which model to use, start with a general-purpose model and only use a different type of model when you don’t like what you get.
There are more types of models (a lot more, actually), but this list covers the most popular ones.
Future posts will look at some of these model categories in more detail. Stay tuned.
Comments