Rate Limits You May Encounter When Using the Gemini Text Agent

When using the Gemini Text Agent action in Zenphi, you can use either your own Gemini API key or Zenphi's API key through Zenphi Add-ons.

Using Your Own Gemini API Key

If you're using your own Gemini API key, your requests are subject to Google's API rate limits. These limits depend on your Google Cloud project, the Gemini model you're using, and your usage tier.

The main limits are:

  • RPM (Requests Per Minute): The maximum number of API requests you can make per minute.

  • TPM (Tokens Per Minute): The maximum number of input tokens you can process per minute.

  • RPD (Requests Per Day): The maximum number of API requests you can make per day.

If you reach any of these limits, your requests may fail with a 429 RESOURCE_EXHAUSTED error.

You can check your current rate limits in Google AI Studio:

https://aistudio.google.com/

Go to Dashboard → Rate Limits and check the limits for the Gemini model you're using.

Please note that these limits generally apply at the Google Cloud project level, not separately to each API key. If you have multiple API keys under the same project, their usage typically shares the same quota.

Using Zenphi Add-ons

Alternatively, you can use Zenphi's API key through Zenphi Add-ons instead of your own Gemini API key.

The good news is that you won't be subject to your own Google Gemini API project's rate limits when using Zenphi's API key. Zenphi manages the underlying API access, so you don't need to monitor or manage your own Gemini API quota.

Zenphi's own usage limits and add-on allowances still apply, but you may be able to run significantly more requests before encountering rate limits.

Which Option Should You Choose?

If your workflows make only a small number of Gemini requests, using your own API key may be sufficient.

However, if you're planning to use Gemini heavily or run workflows that generate many requests in a short period, using Zenphi Add-ons can be a more convenient option. It reduces the need to manage your own API limits and may help you avoid hitting your project's rate limits during heavy usage.