I’m on the free tier. I’m using Gemini 2.5 flash. I have 250 requests per day and used only 131..
I’ve sent more than 10 requests per minute and got an error for that, and that’s perfectly understandable, but hours passed after that, and I still get the limit reached error.
This looks like a bug; it considers me reaching the RPM, as if I reached the limit for the whole day.
Have a look at the screenshot; the RPM limit is permanent. Shouldn’t it be clear after 1 minute passes?
Update:
Things got even stranger. I tried using my other API key, and it worked fine. So the restriction goes only for this particular API.
Aren’t the restrictions based on a project? Not the individual API.
This shows that it’s definitely a bug on Gemini’s side.
I never get an answer in this community. Can I get a response here, please? Don’t leave this topic empty, for God’s sake, this problem exists, and I want to hear what’s going on.
I’m with this exactly same error and i just can’t find anyone who has actually solved that. I’m also want to know what’s happening because this is driving me crazy
Hello @aroshidze & @Marcus_Conceicao,
If you continue to experience this issue, please provide the following details to assist us in our investigation:
Additionally, it would be beneficial if you could briefly describe the task you are using the API for, or provide the exact prompts used.
Complete Erro Message:
{
“error”: {
“code”: 429,
“message”: "You exceeded your current quota, please check your plan and billing details. For more information on
this error, head to: https:// ai.google.dev/gemini-api/docs/rate-limits. To monitor your current usage, head to: htt
ps://ai.dev/usage?tab=rate-limit. \n* Quota exceeded for metric: generativelanguage .googleapis.com/generate_content_
free_tier_input_token_count, limit: 0, model: gemini-3-pro\n* Quota exceeded for metric: generativelanguage.googleap
, limit: 0, model: gemini-3-pro\n* Quota exceeded for metric: generativela
nguage .googleapis.com/generate_content_free_tier_requests, limit: 0, model: gemini-3-pro\n* Quota exceeded for metri
c: generativelanguage .googleapis.com/generate_content_free_tier_input_token_count, limit: 0, model: gemini-3-pro\nPl
ease retry in 49.003386475s.",
“status”: “RESOURCE_EXHAUSTED”,
“details”: [
{
“@type**”: “type .googleapis.com/google.rpc.Help”@type
“links”: [
{
“description”: “Learn more about Gemini API quotas”,
“url”: “https:// ai.google.dev/gemini-api/docs/rate-limits”
}
]
},
{
@type “@type”: “type .googleapis.com/google.rpc.Quota**@typeailure”,
“violations”: [
{
“quotaMetric”: “generativelanguage .googleapis.com/generate_content_free_tier_input_token_count”,
“quotaId”: “GenerateContentInputTokensPerModelPerDay-FreeTier”,
“quotaDimensions”: {
“location”: “global”,
“model”: “gemini-3-pro”
}
},
{
“quotaMetric”: “generativelanguage .googleapis.com/generate_content_free_tier_requests”,
“quotaId”: “GenerateRequestsPerDayPerProjectPerModel-FreeTier”,
“quotaDimensions”: {
“model”: “gemini-3-pro”,
“location”: “global”
}
},
{
“quotaMetric”: “generativelanguage .googleapis.com/generate_content_free_tier_requests”,
“quotaId”: “GenerateRequestsPerMinutePerProjectPerModel-FreeTier”,
“quotaDimensions”: {
“model”: “gemini-3-pro”,
“location”: “global”
}
},
{
“quotaMetric”: “generativelanguage .googleapis.com/generate_content_free_tier_input_token_count”,
“quotaId”: “GenerateContentInputTokensPerModelPerMinute-FreeTier”,
“quotaDimensions”: {
“location”: “global”,
“model”: “gemini-3-pro”
}
}
]
},
{
@type “@type**”: “type .googleapis.com/google.rpc.Retry**@typenfo”,
“retryDelay”: “49s”
}
]
}
}
Biling Tier: Free
Model Name: it’s happening in gemini 2.5 pro and gemini 3 pro
Region: Brazil
Plataform: Gemini API SDK
I’m using on a Agentic AI project
Hello,
Based on the error log you shared, it appears you are hitting four distinct quota violations:
This is likely occurring because your Agent is executing multiple API calls in rapid succession. Please note that the Free tier has stricter rate limits.
You can monitor your API usage on the AI Studio Dashboard to track your current consumption. To prevent these errors, we recommend adding sufficient waiting time between your API calls.
