Inference API cost changed for meta-llama-3.3-70b?

For the meta-llama-3.3-70b model, I had made about 170 requests on 20th/21st March and it had used up $0.02 of my $2 free limit. For the same model and similar amount of tokens, when I made about 30 requests on 2nd April, it charged me $0.50 instead. The first 170 requests are labelled “paid” and the recent 10 say “Pending” . My subscription started from 20th March so it should be valid until 20th April, right? What is going on? Why did a new period start from the 1st of April?

1 Like

To be honest, I don’t think any of the users understand. I don’t even know how much the staff understand… @julien-c

In February, Inference billing usage had been a fixed rate while we added pay-as-you-go billing support. Starting in March, usage now takes into account compute time x price of the hardware. We’re really sorry for any confusion!

We have more information about Inference Providers here: Inference Providers.

1 Like

Hey, thanks so much for the response!

Still a question, though - if I paid for the subscription at 20th March, 2025, why was my subscription period until 1st April and not 20th April? Also, if the change was made in March, why was I still billed lower at first - when exactly was this change implemented?

Thank you for your support in advance.

1 Like