RATE LIMITS
Generous by default. Scalable on demand.
Per-User Limits
| MODEL | RPM | TPM | RPD |
|---|---|---|---|
 GPT-5.4-Mini | 30 | 50000 | 1500 |
| GPT-5.4-Nano | 50 | 100000 | 3000 |
| GPT-5.3 | 20 | 30000 | 1000 |
| GPT-5.2 | 20 | 30000 | 1000 |
| GPT-5-Mini | 30 | 50000 | 1500 |
| GPT-5-Nano | 50 | 100000 | 3000 |
 Gemini-3.1-Pro | 15 | 50000 | 800 |
| Gemini-3-Flash | 30 | 50000 | 1500 |
| Gemini-3.1-Flash-Lite | 50 | 50000 | 1500 |
| Gemini-2.5-Pro | 20 | 50000 | 800 |
 Kimi-K2.5 | 20 | 20000 | 800 |
 Grok-4.1-Fast | 30 | 50000 | 1500 |
| Grok-4.1-Fast-NR | 30 | 50000 | 1500 |
 Mistral-Large-3 | 20 | 30000 | 1200 |
 DeepSeek-V3.2 | 15 | 20000 | 800 |
RPM = requests per minute. TPM = tokens per minute. RPD = requests per day.
Educational Discount
Student builder? Contact us for an educational discount and free credits on us.
Need Higher Limits?
The defaults above cover most use cases. If you need more headroom for a high-volume project, reach out and we'll sort it out.