Free
$0/ mo
- 10,000 requests / month
- 60 requests / minute
- Single API key
- Community support
pricing
Free for prototypes.
Cheap for production.
One model. One endpoint. Two prices. We bill metered after the free tier — pay for what you actually classify.
recommended
Pro
$9/ mo + $0.0001 / req
- 1,000,000 requests / month included
- 600 requests / minute, burst-tolerant
- Unlimited keys
- Email support, 1 business day
Both plans share the same model and the same /v1/predict reference. Need higher volume, dedicated capacity, or self-hosted inference? Email hi@siftfy.io — we'll work it out.