Hugging Face Inference Endpoints SLA
Hugging Face · AI models · no public SLA · checked 25h ago
No public SLA, so there is no promise to test.
Incidents, last 365 days
No public status history.
Commitments and credits
Docs FAQ gives only availability guidance, not a commitment: 'To help mitigate service interruptions on an Inference Endpoint that needs to be highly available, please make sure to use at least 2 replicas'. Status page shows observed (not committed) 90-day uptime, e.g. 'Inference Endpoints UI 99.924% uptime' on 2026-09-23.
More in ai models
ServicePromisedFirst creditObserved 365dDowntime
Amazon BedrockAWS · AI models · Standard99.9%First credit10%Observed 365d99.178%All incidents 99.852%
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%
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
- No uptime commitment published.
- Not covered
- Services provided 'as is' and 'as available'; Single-replica or scale-to-zero endpoints may be interrupted; No public service-credit schedule
- Status history
- No public status history.
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.