Managing OpenAI Cost
GPT-3 has different engines available with different levels of accuracy. In the preferences, you can choose "Highest", "Medium", or "Low". By default, it uses the highest accuracy level. You can adjust that in the Preferences under "Cost & Accuracy"
Highest
Highest uses Davinci. Davinci is the most capable model family and can perform any task the other models can perform and often with less instruction.
$0.0200 / 1K tokens
Medium
Uses the Curie engine. Very capable, but faster and lower cost than Davinci.Curie is extremely powerful, yet very fast.
$0.0060 / 1K tokens
Low
Uses the Ada engine. Ada is usually the fastest model and can perform tasks like parsing text, address correction, and certain kinds of classification tasks that don’t require too much nuance. Ada’s performance can often be improved by providing more context.
$0.0008 / 1K tokens
As you can see, "Medium" type is 10 times cheaper than the "Highest". Depending on your use case, you can choose the best option.
Comments
Your comment has been submitted