AI Quick Tips 231: AI Flash Models

When you start paying attention to the models that you are using with AI tools, you’ll eventually come across flash models.  These are smaller, faster models that can’t do as much as some other models but you may want to use them anyway.

Most of the popular AI chatbots - ChatGPT, Gemini, Claude, Copilot, etc. have flash models.  You won’t always have the option to choose them though.  Sometimes that choice is made for you behind the scenes.  

When you do get to choose in the chatbots (or when you start using the APIs), here are some tasks where flash models can be used - 

• Short-form writing
• Brainstorming (shorter, simpler idea generation)
• Quick explanations
• Customer support assistants (most customer requests are short answers)
• Text summarization

Pros of flash models -

• Lower memory usage (take fewer computing resources so they can run on less powerful devices - think cell phones)
• Lower cost 
• Faster
• Less energy usage
• If the task is simple, you would never know the difference

Cons of flash models - 

• Small context window
• Not good at thinking/reasoning
• Higher rate of hallucinations
• Less creative responses
• Not good at complex tasks

Like all of the other types of models, there is definitely a place for flash models.  As you use AI tools more, you’ll start to see when using flash models will make sense (and cents) for you.