Kurrens Service Level Agreement
Last Updated: September 28, 2026
This Service Level Agreement (“SLA”) applies to customers of INFERERA PTE. LTD. (“Kurrens”) who have an Order Form that references this SLA (“Eligible Customers”). It forms part of the Terms of Service. It does not apply to free usage, Promotional Credits, Flex service tier requests, batch jobs, or models labeled Preview or Deprecated.
1. Service Commitment
Kurrens will use commercially reasonable efforts to make the Serverless API available with a Monthly Uptime Percentage of at least [[99.9%]] for each covered model. Dedicated endpoints are covered at the percentage stated in the Order Form.
2. Definitions
- “Valid Request” means an API request from an Eligible Customer to a covered model that is properly authenticated, well-formed, within the Customer’s rate limits and balance, and not blocked for violating the Acceptable Use Policy.
- “Failed Request” means a Valid Request that returns a 5xx status code, fails mid-stream, or does not return a first token within [[60]] seconds.
- “Excluded Responses” means 400, 401, 402, 403, 404, 413, and 429 responses caused by the Customer’s request, credentials, balance, or rate limits.
- “Monthly Uptime Percentage” means, for a calendar month, (Valid Requests − Failed Requests) ÷ Valid Requests × 100, calculated per model, over 5-minute intervals with at least [[100]] Valid Requests. Intervals with fewer requests are excluded.
3. Service Credits
| Monthly Uptime Percentage | Service Credit (% of fees for the affected model in that month) |
|---|---|
| < [[99.9%]] and ≥ [[99.0%]] | [[10%]] |
| < [[99.0%]] and ≥ [[95.0%]] | [[25%]] |
| < [[95.0%]] | [[50%]] |
Service Credits are applied as account credits against future usage. They are not refundable or payable in cash. Total Service Credits for a month may not exceed [[50%]] of the fees paid for the affected model in that month.
4. Claiming Credits
Send a claim to support@kurrens.ai within thirty (30) days after the end of the month in which the shortfall occurred, including the affected model, the time periods, and request IDs (which are included in every API response). We will confirm eligibility based on our request metadata.
5. Exclusions
This SLA does not apply to unavailability caused by: (a) scheduled maintenance announced at least [[48 hours]] in advance on https://status.kurrens.ai (not exceeding [[4]] hours per month); (b) factors outside our reasonable control, including force majeure, internet outages outside our network, or third-party attacks we could not reasonably prevent; (c) the Customer’s systems, integrations, or misuse; (d) suspension under the Terms of Service; or (e) models or features labeled Preview, Beta, or Deprecated.
6. Sole Remedy
Service Credits are the Customer’s sole and exclusive remedy for any failure to meet this SLA, except as otherwise stated in an Order Form.
7. Status and Incident Communication
We publish real-time status and incident history at https://status.kurrens.ai. Incidents are communicated through the stages Investigating, Identified, Monitoring, and Resolved. Customers can subscribe to email notifications.