Operations
Rate limits
Two dimensions: requests per minute (smooth-out) and tokens per month (plan quota). Both return 429 when exceeded.
Per-plan limits
| Plan | Requests / min | Tokens / month | Max API keys |
|---|---|---|---|
| Free | 60 | 100,000 | 2 |
| Pro ($19/mo) | 300 | 10,000,000 | 10 |
| Team ($49/mo) | 1,000 | 100,000,000 | Unlimited |
| Enterprise | Custom | Unlimited | Unlimited |
Requests per minute
Rate limiting is applied per API key using a sliding-window counter. When exceeded, you get 429 Too Many Requests with a Retry-After header indicating how many seconds to wait.
Monthly token quota
The quota counts output tokens from your compressed responses (i.e., what mintoken saved you from paying for at the provider, but counts against your mintoken plan). When approaching the limit, every response carries:
| Header | Value |
|---|---|
X-Mintoken-Tokens-Used | Current cumulative token usage this billing cycle. |
X-Mintoken-Tokens-Limit | Your plan's monthly limit. |
When you exceed the quota, subsequent requests return 429 token quota exceeded until the next billing cycle starts.
Zero-spend protection
Mintoken never auto-charges you for overage. Hitting your limit fails fast — upgrade in the dashboard to raise the ceiling. This is by design: no surprise end-of-month bills.
Request size
Every plan has a per-request body-size cap to protect the proxy. Exceeded: 413 Payload Too Large.
| Plan | Max request body |
|---|---|
| Free | 512 KB |
| Pro | 2 MB |
| Team | 8 MB |
| Enterprise | Negotiable |