Rate Limits
Rate limits are enforced around account tier, API key settings, input tokens per minute, output tokens per minute, and request count per minute.
Default developer API limits
Section titled “Default developer API limits”| Tier | Input tokens/min | Output tokens/min | Requests/min |
|---|---|---|---|
beginner | 500,000 | 80,000 | 60 |
steady | 2,000,000 | 200,000 | 120 |
persistent | 5,000,000 | 400,000 | 300 |
tryharder | 10,000,000 | 800,000 | 1,200 |
| unlimited override | 100,000,000 | 50,000,000 | 100,000 |
These defaults can be overridden by environment configuration.
Key-level caps
Section titled “Key-level caps”API keys can be account-pool keys or custom-limited keys. Custom keys may reduce, but not exceed, the account tier limits for:
- input tokens per minute
- output tokens per minute
- requests per minute
- maximum input tokens per request
- maximum output tokens per request
- allowed model list
Enforcement
Section titled “Enforcement”Quota is reserved before request processing and final usage is settled after the response completes. For generation endpoints, SHINE SHOP DEV estimates input/output usage from the request and updates settlement from final usage metadata when available.
Account usage history is available in the web panel and from /api/account/usage.