Roo-Code: Roo Code does not always adhere to rate limits

App Version

3.14.0

API Provider

Google Gemini

Model Used

gemini-2.5-flash-preview-04-17:thinking

Actual vs. Expected Behavior

I want to use the free tier of gemini-2.5-flash-preview-04-17:thinking, so i put in a free tier API key and set the rate limit to 7 seconds as the official rate limit for the free tier of this model is 10 RPM: https://ai.google.dev/gemini-api/docs/rate-limits#free-tier And sometimes i see the countdown between requests and everything works fine but sometimes, especially after reading a file, the next api request is executed right away without countdown and i run into 429 errors. Why is that? Please fix it, i will also try to. Cheers Jay

Detailed Steps to Reproduce

  1. set API provider Google Gemini
  2. set free tier API key from https://aistudio.google.com/
  3. set Rate limit slider in Roo Code settings to 7s

Relevant API Request Output

got status: 429 Too Many Requests. {"error":{"message":"exception parsing response","code":429,"status":"Too Many Requests"}}

Additional Context

No response

About this issue

  • Original URL
  • State: closed
  • Created 2 years ago
  • Reactions: 7
  • Comments: 20

Most upvoted comments

Facing same issue ,

This makes roo code basically pointless for me

I am sorry i closed this as completed by accident. It is not afaik.

I used to use Cline, and it started to give 429 at every prompt, even with an API not used. Then I started to use Roo, but still, after a few requests the 429 appears again. It would be good to be able to store more than one api to rotate 😃

thats a nice idea!

The rate limit feature is absolutely useless like this.