Nano Banana(Free) is being rate limited

[unique sage]

Account with over $20 in credits. Today I started getting 429 while calling google/gemini-2.5-flash-image-preview:free

{
"error": {
"message": "Provider returned error",
"code": 429,
"metadata": {
"raw": "google/gemini-2.5-flash-image-preview:free is temporarily rate-limited upstream. Please retry shortly, or add your own key to accumulate your rate limits: https://openrouter.ai/settings/integrations",
"provider_name": "Google AI Studio"
}
},
"user_id": "user_2(...)"
}

I understand this is not an OpenRouter limitation, but a Google new temporary (hopefully) restriction. I've also set up my own key in integrations, but I still get the same errors. When using my own AI Studio Keys, should I still use the :free version?

Or I should drop the free entirely and pay for generations?

[charred meadow]

I'm an AI assistant - my answers may be incorrect and I cannot see screenshots. Please verify information with the OpenRouter docs.

If you have added your own AI Studio key in OpenRouter integrations and want to use your own quota, you should not use the :free version of the model. Instead, use the base model identifier (for example, google/gemini-2.5-flash-image-preview) without the :free suffix. The :free version is specifically routed through OpenRouter's shared pool, which is subject to upstream rate limits and quotas.

Quoting from the error message:
"Please retry shortly, or add your own key to accumulate your rate limits: https://openrouter.ai/settings/integrations"

• Switch your API calls to the base model name (drop the :free) after setting your own key.
• This will route calls using your own Google AI Studio quota and avoid shared pool limits.

You can find and manage your integrations at OpenRouter Integrations.

If you continue to have issues after switching, please contact the mods through the Discord server for further support.

I don't have specific documentation for this exact workflow, so please verify with the mods or check the OpenRouter docs directly for the latest best practices.

-# This is an automated response powered by Answer HQ. Use !mute to stop the bot from responding.

[unique sage]

So, even with my own Gemini API keys, there's no way to access the free model at this time. Correct?

[charred meadow]

That’s correct—if you use your own Gemini API keys, you should NOT use the :free models. The :free models on OpenRouter are routed through OpenRouter’s shared quota, not your personal key. When you add your own key, you must use the base model identifier (without :free) to access the paid/quota-based version with your own limits.

The free (:free) models are only accessible via OpenRouter’s shared pool, and if they're rate-limited upstream, using your own key won't help. To use your own quota, drop the :free suffix from the model name.

I don't have specific documentation for this exact scenario. Please verify with the mods or check the OpenRouter docs for confirmation.

-# This is an automated response powered by Answer HQ. Use !mute to stop the bot from responding.

[slate mesa]

Account with over $20 in credits. Today I started getting 429 while calling google/gemini-2.5-flash-image-preview:free