Hack Club AI

Appearance

Rules & Rate Limiting ​

Rules ​

  1. For teens. This service is for teens 18 and under only. Hack Club is a charity — please do not abuse this service.

  2. No coding agents. You are not allowed to use this service with coding agents like Cursor.

  3. No proxies. You are not allowed to use this service to create proxies or other tools that allow others to access the API without them also abiding by these rules.

  4. No resale. You are not allowed to resell this service or use it to create a service that resells AI to others.

  5. Don't use this service with OpenClaw or similar tools like Hermes Agent or NanoClaw. They waste tokens and provide little benefit.

  6. Follow the Code of Conduct. You are not allowed to use this service to create tools that intentionally violate the Code of Conduct. Don't try to generate explicit imagery or text, malware, or other harmful content.

Daily Spending Limit ​

Every account can spend $3 per day across all endpoints. The day resets at 00:00 UTC.

Each request reserves its estimated cost up front and is charged what the provider actually reports when it finishes, so you aren't charged for requests that fail upstream. Requests that would take you over the limit get a 402 Payment Required response. Need a higher limit? Email hey@mahadk.com.

Your spend for today is shown at the top of your dashboard.

Rate Limiting ​

Every user can make 7,500 requests per 30 minutes, shared across all endpoints.

When you exceed the rate limit, the API returns 429 Too Many Requests with a Retry-After header saying how many seconds to wait.

Errors ​

Errors are returned as JSON with an error message:

json
{ "error": "Rate limit exceeded. Retry in 42s." }
StatusMeaning
400The request body is invalid (not JSON, or missing a required field).
401The API key is missing or invalid.
402You've reached your daily spending limit.
403Your account is banned, needs identity verification, or the request was blocked by the rules above.
429You've hit the rate limit. Wait for the time in Retry-After.

Errors from the upstream model provider (for example, an unknown model ID) are passed through with the provider's status code and body.