#Perplexity: Llama3 Sonar 8B Online - Cost Issue

4 messages · Page 1 of 1 (latest)

violet copper
#

I am using perplexity/llama-3-sonar-small-32k-online for my app, but the price calculated on activity page seems way off.

Price mentioned for this model is $0.2/M input tkns. $0.2/M output tkns

I got charged for 64 input token + 28 output token = $0.00497

But calculation says that it should be $0.0000184. It is 270X more than what I am calculating. Is there something wrong with my calculation or there is some other issue. Please help this is a blocker for me to go live with latest version of my app

Screenshot_2024-06-17_at_23.26.33.png Screenshot_2024-06-17_at_23.26.16.png
tight vortex
#

@violet copper Did you account for the flat per-request fee for online perplexity models?

image.png
violet copper
#

How does flat per-request fee that work?

tight vortex
#

It mean that every request costs $0.005, in addition to the token pricing, since you're also paying for the online functionality