Amazon Bedrock SLA
AWS · AI models · promises up to 99.9% · SLA of Oct 4, 2023 · checked 25h ago
3 major outages in the last 365 days, each in one region (me-south-1, me-central-1 and us-east-1).
Worst region me-central-1: 72 h counted, 99.178% observed against the 99.9% Standard, which applies per region. 2 of 12 months fell below it (credits are claimed per month). 2 incidents ran past 72 hours; each counts for 72 hours at most (see the method).
Last 90 days: no major outage.
Incidents, last 365 days
Major outage time per month
Allowed per month: 44 min
Commitments and credits
- StandardAmazon Bedrock APIs for models available on Bedrock, per AWS region99.9%below 99.9%: 10% · below 99%: 25% · below 95%: 100%
More in ai models
Claude API (Standard tier)Anthropic · AI modelsNoneFirst creditn/aObserved 365d98.76%All incidents 94.774%
Cohere API (SaaS: Command, Embed, Rerank)Cohere · AI modelsNoneFirst creditn/aObserved 365d100%All incidents 99.986%
Vertex AI (Gemini API on Vertex)Google Cloud · AI models · Vertex AI Platform: Training, Deployment, Batch Prediction99.9%First credit10%Observed 365d100%All incidents 99.488%
Hugging Face Inference EndpointsHugging Face · AI modelsNoneFirst creditn/aObserved 365dNo feed
Azure OpenAI in Foundry ModelsMicrosoft Azure · AI models · Foundry Models (Sold by Azure), availability99.9%First credit10%Observed 365dNo feedAll incidents 100%
Terms, exclusions and sources
- Measured
- Average of 5-minute Availability (share of Bedrock API requests not returning a 500 error) per AWS region
- Credit cap
- Up to 100% of Bedrock charges in the affected region at the lowest tier; issued only if over $1
- Claim window
- By the end of the second billing cycle after which the incident occurred
- How to claim
- Open a case in the AWS Support Center with 'SLA Credit Request' in the subject line and request logs
- Not covered
- Factors outside AWS's reasonable control; Customer actions or inactions (e.g. provisioned throughput scaling, VPC/credential misconfiguration); Not following Bedrock documentation best practices; Customer equipment, software or other technology; Underlying software causing repeated model crashes or an inoperable model
- Status history
- Status history since Sep 23, 2025
- Weighted uptime
- All incidents, weighted: 99.852% · Downtime
Monthly Uptime Percentage ... Less than 99.9% but greater than or equal to 99.0% 10% ... Less than 99.0% but greater than or equal to 95.0% 25% ... Less than 95.0% 100%
The commitment is expressed only via the credit table ('Monthly Uptime Percentages set forth in the table below'); 99.9% is the table's top threshold. No latency/throughput SLA is published.
Sources: each vendor's own SLA page (re-read monthly) and status page feed (read every Friday). A summary, not the contract: the contract you sign governs. Logos via logo.dev; trademarks belong to their owners.