Expert.ai NL API Edge Service Level Agreement
The below terms govern the service levels applying to the expert.ai NL API Edge Service that expert.ai has agreed to provide to Customer under the Special Conditions of Service, its Annexes and the General Terms of Service.
During the term of the Service Contract, provided the Customer meets the below rate limits, expert.ai guarantees, with reasonable efforts, a Service availability uptime of 99.5% on a monthly basis.
Current request rate limits for the expert.ai Natural Language API are as follows:
Maximum API requests per minute: 180
Maximum API requests per day: 500
Should expert.ai fail to meet the above availability threshold, while Customer meets his obligations under the SCS, Customer will be eligible to receive financial credits described below, as a sole and exclusive remedy.
Financial credits shall be calculated as in the table below on the basis of the observed downtime of the service, calculated as the number of full minutes during which more than 5% of valid API requests are answered with an HTTP Status 500, divided by the number of minutes in a month. Periods of less than a minute will not be included in the observed downtime.
|Monthly uptime||Percentage of monthly bill applied as credit|
|99.5% to 99%||10%|
|Between 99% and 95%||20%|
|Less than 95%||50%|
Total financial credits will not exceed 50% of the amount due by the Customer for the applicable month. Financial credits will be made in the form of monetary credits applied to future use of the Service and must be applied within 60 days after the credits were attributed.
In order to receive financial credits, Customer must notify expert.ai by email at email@example.com within 30 days from the time of the observed downtime and provide date and time such errors occurred.
The SLA does not apply to (i) features designated as Beta in the associated documentation, or to (ii) the API Free Tier, or to (iii) scheduled maintenance operations for which expert.ai will provide Customer 3 business days' advance notice or to (iv) errors caused by factors outside of expert.ai's reasonable control, including but not limited to failures arising from third party systems (such as connectivity services, hosting services or software), failures arising from Customer's software, hardware, or other systems, failures arising from behaviors violating the GTS or SCS, or arising from Customer's use of the Service in a manner that is inconsistent with the documentation, or by quotas being applied by the API Service.