Open W-Lawless opened 1 year ago
The rate limit is 3 requests a second on the free trial, and 60 if you are a pay as you go customer.
| TEXT & EMBEDDING | CHAT | EDIT | IMAGE | AUDIO -- | -- | -- | -- | -- | -- Free trial users | 3 RPM150,000 TPM | 3 RPM40,000 TPM | 3 RPM150,000 TPM | 5 images / min | 3 RPM Pay-as-you-go users (first 48 hours) | 60 RPM250,000 TPM | 60 RPM60,000 TPM | 20 RPM150,000 TPM | 50 images / min | 50 RPM Pay-as-you-go users (after 48 hours) | 3,500 RPM350,000 TPM | 3,500 RPM90,000 TPM | 20 RPM150,000 TPM | 50 images / min | 50 RPMhttps://platform.openai.com/docs/guides/rate-limits/overview
That's odd, I specifically upgraded to a paid plan so I could get an API key for this library.
Please check that this issue hasn't been reported before.
Expected Behavior
The application should run the example script. I checked openAPI and the api rate limit is three requests per minute? Is this a recent change? Seems like they are intentionally trying to break people's scripts/use of automated agents with a rate limit that low...
Current behaviour
Console reports 'rate limit reached trying again in 20 seconds'
Steps to reproduce
Clone repo Create .env with API key run python3 examples.py file
Possible solution
Rate limit the API calls with set timeout ?
Which Operating Systems are you using?
Python Version
LoopGPT Version
latest
Acknowledgements