Intelligence VS Cost-per-Task LLM Comparison

OpenAI

Heat trend

Collecting trend data

The percentage is based on available heat signal, not comment count or independent people.

Using Artificial Analysis as the guide for cost per task and intelligence index, I was able to generate a few graphs of the latest models and compare them. The graphs are a bit hard to follow, but here is the order:

Tell me what you guys think. I used Gemini for the graph generator and to retrieve the data. Then I used Claude to double-check the scores, prices, and placements on the graphs were correct.