Skip to main content
Watch: I Built an AI App Without Leaving Databricks

Google Gemini API error

Gemini API 503 UNAVAILABLE: what it means and how to fix it

Error message
503 UNAVAILABLE

Retry? YesLast verified 3 documented fixes2 sources

What it means

generateContent API: the service may be temporarily overloaded or down.

Causes

  • The service is temporarily running out of capacity

How to fix it

  1. Fix 1.

    Check the Gemini API status page for ongoing incidents.

  2. Fix 2.

    Temporarily switch to another model (for example from Gemini 2.5 Pro to Gemini 2.5 Flash).

  3. Fix 3.

    Wait and retry with exponential backoff. If it persists, report it with the Send feedback button in Google AI Studio.

More Gemini API errors

All 30 Gemini API errors

HTTP 503 at other providers

Compare 503 across providers

FAQ

What does the Gemini API error "503 UNAVAILABLE" mean?

generateContent API: the service may be temporarily overloaded or down.

How do I fix Gemini API "503 UNAVAILABLE"?

There are 3 documented fixes. Start with the first: Check the Gemini API status page for ongoing incidents.

Should I retry Gemini API "503 UNAVAILABLE"?

Yes. The official docs treat this error as retryable.

Which other AI APIs return HTTP 503?

Per their official error docs: Amazon Bedrock (503 ServiceUnavailableException), Azure OpenAI (503 Server Error), DeepSeek API (503 - Server Overloaded), Groq API (503 Service Unavailable), Mistral API (503 Service unavailable), OpenAI API (503 service_unavailable_error server_is_overloaded), OpenRouter (503 No available model provider meets your routing requirements; 503 provider_overloaded), Vertex AI (503 UNAVAILABLE).

Sources

Last verified against 2 sources: Gemini API's official docs and official SDK repositories.

Newsletter

Know when the APIs change

One email per week. New model releases, API changes, what broke, and how developers are working around it.