Rate limits for new embedding v3 models

Are there any official statements on the rate limits for the new embedding v3 models (text-embedding-3-small and text-embedding-3-large)? https://platform.openai.com/docs/guides/rate-limits/usage-tiers seems not yet to mention these models.

On https://platform.openai.com/account/limits I can see that my personal account has the same rate limits for them as for text-embedding-ada-002, but I would like to know the general rate limits for other users. Can we expect that for all users in all tiers, rate limits for text-embedding-3-* are the same as for text-embedding-ada-002?

1 Like

I’ve yet to hit a rate limit.

However, I’ve had lots of these:

OpenAI HTTP Error (spotted in ruby-openai 6.3.1): {"error"=>{"message"=>"The server had an error while processing your request. Sorry about that! You can retry your request, or contact us through our help center at help.openai.com if the error persists. (Please include the request ID 243e9sadfjlsadkfjlasdkfjlsadkfj in your message.)", "type"=>"server_error", "param"=>nil, "code"=>nil}}

which are very annoying!

Hi, that doesn’t look like a rate limit but more like an internal error.

I agree with you, that was kind of my point.

1 Like

Rate limits for embedding-3 models are now documented for all usage tiers:


However, they have removed the rate limit documentation for the old ada-002 models in turn …

1 Like