Inquiry regarding API rate limits – Credits available on accountTier:

[hmegcode]

Current Tier

Spore (0.01 pollen/hour)

Question

Dear Pollinations Support Team,

I am writing to you regarding some technical limitations I am encountering with your API.

Recently, I purchased $100 worth of Pollen, and thanks to your promotional offer, my account was credited with $200. My goal was to use these credits to power various community websites and projects by distributing my API keys.

However, despite having a significant balance, we are frequently hitting rate limits. This is preventing the community from using the tools effectively.

Since I have invested in a large volume of Pollen to support multiple projects simultaneously, I would like to request a limit increase or the removal of these restrictions for my account. Without this adjustment, it is difficult to see the utility of purchasing a high volume of credits if they cannot be utilized across various platforms as intended.

Could you please look into my account and let me know if it is possible to lift these constraints?

Account Details:

Email: [Votre adresse email associée au compte]

Current Balance: ~$194 Pollen

Thank you in advance for your help and for the great work you do with Pollinations.

Best regards,

Hicham

[ElliotEtag]

Hi @hmegcode — sorry we couldn't get back to you earlier on this.

Quick context first: every Pollinations account gets the same maximum rate limits regardless of tier or pack balance. The throttling you're hitting is coming from a downstream provider, not from us, so to reduce it we need to know exactly which model is being slowed.

Could you share:

1. Which model(s) are giving you rate-limit errors? (flux, gpt-5.4-nano, gemini-2.5-flash-lite, etc.)
2. Roughly how many requests per minute (or per second) you're trying to send when you see the throttling.
3. The exact error response body and HTTP status code you're getting back from the API (a copy-paste of one is ideal).
4. Are the requests originating from a single key, or fanned out across many keys / community apps?

With that, we can identify the bottleneck on the provider side and see what can be done. Thanks.