✨ Claude vs. GPT: which is better for note-taking?

There is a battle between LLMs right now, and it can be hard to know which one to use.

Anthropic’s Claude 3.5 Sonnet seems to be winning right now, but not with everything. Fortunately, it’s pretty easy to toggle between the two right from within your notes.

This email will show you how to do just that, and then show some examples of the differences between the two.

Toggling between Anthropic and OpenAI LLMs

Reflect lets you switch back and forth between GPT-4o and Claude 3.5 Sonnet from the preferences page. By default it will be set to Anthropic’s Claude.

You are also able to add API keys for each, which will let you get unlimited usage based on your billing preferences with the respective provider.

Performance metrics of each

Below are some performance benchmarks across the major LLMs:

As you can see, Claude 3.5 Sonnet is the clear winner, but there are still some things GPT-4o does slightly better.

Side by side comparison

Ultimately, the best way to see the difference between the two is to view them both in action.

I’ve done a side-by-side comparison in this note where I’ve run the same prompt on the same sample text using both Claude and GPT. You can see the difference directly alongside the original.

Finally, remember that we collect custom AI note-taking prompts in this spreadsheet.

/