The error 429 say that try again in 7m12s. but it is no use. M tier free. and how to define “a day”? from my last use the api or from the local time?
Welcome to the forum…
The HTTP status code 429 indicates that you’ve reached the rate limit for your API requests. The message “Try again in 7m12s” suggests that you should wait for 7 minutes and 12 seconds before making another API request. However, the behavior of rate limits and how they reset can vary depending on the API provider’s policies.
In the case of OpenAI’s GPT-3.5 Turbo API, the rate limits are typically defined in terms of Requests Per Minute (RPM) and Tokens Per Minute (TPM). For the free “M” tier, the rate limits are as follows:
| Model | RPM | RPD | TPM | TPD |
|---|---|---|---|---|
| gpt-3.5-turbo | 3 | 200 | 20,000 |
If you reach these limits, you’ll need to wait until the rate limit resets before making more requests. The rate limits usually reset on a rolling window basis. This means that you don’t have to wait until the end of a calendar day; instead, you need to wait for the specified time interval from your last request.
In your case, if the error message says “Try again in 7m12s,” it means you should wait for 7 minutes and 12 seconds from the time of your last API request before making another request.
To avoid rate limit issues, you can monitor your usage and ensure that your application stays within the defined rate limits. If you consistently need more requests or tokens, you may consider upgrading to a higher-tier subscription that provides higher limits.
Be sure to check the docs for rate limits and usage tiers…