Serverless Real-Time Inference Pricing

Pricing

Serverless Real-Time Inference

Make live predictions in your apps and websites.

Select plan

help me choose

Select cloud

Select region

Select
Loading...

Serverless Real-Time Inference pricing examples

Serverless Real-Time Inference pricing examples

 

Notes:

1. The simple examples above are baseline configurations. Example 2 requires launch charge from cold start.

2. Users may also select “scale to zero” or “very latency sensitive” options, which would change DBU emissions.

3. These 2-hour examples assume 115 minute uptime and 5-minute cooldown.

FAQ

4 GB per concurrent request