Forum Discussion
Limit to embedded token generation
No I haven't been able to find out the limit to the token generation without capacity.
I think there are only a few available for you to test your model before going full scale.
So I guess the best way to go about it is to actually put it to test and see where it fails.
Apart from that, I don't think there is any information available from Microsoft regarding this.
If you find anything else/new on it, please do let me know as well.
Thank you,
Vishesh Jain
Alright, Sure.
This confuses me so much.
In order to generate a token you need to provide an account, yes? As soon as I try to do this I need a license. Even if I purchase capacity, it is still requiring a PRO account for the user that is used when calling the api to generate the token. Am I wrong about this? What am I missing?
Hi ryanjlind,
In order to use embedded you need to buy capacity.
However in order to test your embedded model, before you buy the capacity, Microsoft provides some tokens, provided you have a Pro user account, but there is a limit to those tokens that you can generate, while you are testing your model.
So I wanted to know what is that limit. When I am in my testing phase and I have NOT bought the capacity, how many API calls can I generate?
Does it make it better?
Thank you,
Vishesh Jain
Hello, I have tried this but usage is 0. Something is wrong.
