Issue with 429 Error on Gemini API Despite Staying Within Rate Limits

Gemini API
1

I am encountering an issue with the Gemini Flash 1.5 API, where I receive a 429 (Too Many Requests) error after making only 2-4 requests, despite the documented rate limit being 2000 requests per minute (RPM) for my paid plan.

I have already attempted adding delays between requests, yet the issue persists. Based on the API documentation, I should be well within the allowed rate limits, but the system is still throttling my requests unexpectedly.

Could you please help me understand:

  1. Whether there are any undocumented limitations or restrictions that might be causing this?
  2. If my account or API key is being incorrectly rate-limited?
  3. Any recommended troubleshooting steps to resolve this issue?
4 Likes
3

i’m having the same issue. can’t find any solution. on paid tier using gemini-2.0-flash

1 Like
4

Yup, same issue on my side. Trying to process documents, at a rate of 1 small 1-page PDF document every 5 minutes, and after 3 documents I get 429 error. Also trying to use gemini-2.0-flash

I’m on a paid GCP account, with a billing profile that has many other GCP projects with paid-for resources that all work perfectly fine.

1 Like
6

Same issue :confused:
I’m on a paid plan but always getting 429 Too Many Requests error

2 Likes
8

Hi all,

Are you still facing 429 errors?

1 Like
10

Please follow the below instructions to troubleshoot:

Go to GCP console and click “APIs & Services”. Under Metric, search and select “Generative Language API”.. Under “Quotas & System Limits” tab, check for “Current Usage percentage”..

If it reaches 100%, then you have reached your quota limits and hence the 429 Error.

If you think that there is any discrepancy, please DM me with a clear error message and Project ID to help us investigate further.

1 Like
12

using vertex AI. I am most certainly below 100% and I am stuck at 5 req per min.
I hit that limit very fast with embedding (very small token input)
Quota exceeded for aiplatform.googleapis.com/online_prediction_requests_per_base_model with base model: gemini-embedding
pid: crm-system-8e12d

1 Like
14

Hi @Halid_Rian,

Got you.. Thanks for sharing your PID.. I have escalated your issue to Engineering team.. We will investigate this issue soon.

Appreciate your patience!

1 Like
16

Project name :
Project number

1 Like
18

Having this issue since yesterday, showing the error: Resource has been exhausted.

|:—|:—|:—|

7e9affab-2c2a-44c0-b7e7-4097702bc182
7e9affab-2c2a-44c0-b7e7-4097702bc1821511Ă—869 55.6 KB

1 Like
20

@Krish_Varnakavi1 Same here. It works fine before, just happen these two days suddenly. Why is that ?

1 Like
22

@Krish_Varnakavi1 I am seeing a similar problem. I am trying to migrate from OpenAI batching to Gemini and am seeing the 429 error but am on Tier 1. Tried creating a new API key as well just incase. No luck. My dashboard also seems well under. Its One batch job and ~250 requests inside it.

1 Like