Log in
Log into community
Find answers from the community
View all posts
Related posts
Did this answer your question?
๐
๐
๐
Powered by
Hall
Inactive
Updated 3 months ago
0
Follow
perhaps is a stupid question but can
perhaps is a stupid question but can
Inactive
0
Follow
C
CrisTian
2 years ago
ยท
perhaps is a stupid question .... but ... can anyone explain me the difference of LLm tokens and embeddings token ...?? its something that must improve in my algortims??
L
C
7 comments
Share
Open in Discord
L
Logan M
2 years ago
LLM tokens are the number of tokens sent to the LLM (i.e. text-davinci-003 by default, or you might have specified gpt-3.5-turbo, both from OpenAI)
Embedding tokens are the number of tokens sent to the embedding model (i.e. text-ada-002 from OpenAI by default)
This is important because it tells you the cost from OpenAI to build indexes and answer queries
LLMs:
text-davinci-003 = $0.02/1k tokens
gpt-3.5-turbo = $0.002/1k tokens
gpt-4 = $0.03-$0.06/1k tokens
Embeddings:
text-ada-002 = $0.0004/1k tokens
C
CrisTian
2 years ago
ok ... so the 2 of them it will be charged to my ... jojoj
C
CrisTian
2 years ago
ok
L
Logan M
2 years ago
Yup! Every index you make or query you make will use different amounts of tokens
L
Logan M
2 years ago
But it's pretty cheap for the most part in my opinion
C
CrisTian
2 years ago
perfect... and a standar ... charge a $0.01 for a query that his result its a index of a book of 150pages ... is fine ... what do you think ???
L
Logan M
2 years ago
Yea that sounds about right!
Add a reply
Sign up and join the conversation on Discord
Join on Discord