Rate Limits & Quotas
Requests are limited per tier to ensure fair access. Limits apply per API token.
Per-Endpoint Rate Limits
Each endpoint has a sliding-window rate limit (maximum requests per 60-second window):
| Endpoint | Free (req/min) | Premium (req/min) |
|---|---|---|
| /transcribe | 10 | 60 |
| /translate | 30 | 150 |
| /synthesize | 5 | 25 |
| /transcribe_and_translate | 5 | 25 |
| /translate_and_synthesize | 3 | 15 |
Authentication endpoints are also rate-limited: 3 requests per 60 seconds per IP address.
Daily Request Caps
Maximum requests per day, per model:
| Model | Free | Premium |
|---|---|---|
| ASR | 3 | 500 |
| MT | 5 | 1,000 |
| TTS | 3 | 1,000 |
Monthly Volume Caps
Maximum consumption per calendar month:
| Model | Unit | Free | Premium |
|---|---|---|---|
| ASR | minutes | 30 | 3,000 |
| MT | characters | 25,000 | 2,000,000 |
| TTS | characters | 15,000 | 2,000,000 |
When Limits Are Exceeded
429 Too Many Requests
The response includes a detail field explaining which limit was hit (sliding window, daily cap, or monthly volume).
Need higher limits?
Premium tier provides significantly higher caps. Upgrade via your dashboard or contact us for custom enterprise plans.