Models Health Status
This page displays the real-time health and availability status of all AI models across all ToothFairyAI deployment regions. Health data is fetched live from each regional endpoint and aggregated across providers without exposing individual provider information.
Regional Endpoints
Health data is collected from three independent regional deployments:
| Region | Endpoint | Coverage |
|---|---|---|
| AU | https://ai.toothfairyai.com/models_list | Asia-Pacific (Sydney) |
| EU | https://ai.eu.toothfairyai.com/models_list | Europe (Frankfurt) |
| US | https://ai.us.toothfairyai.com/models_list | United States (N. Virginia) |
Each region operates independently with its own infrastructure and health metrics.
Intelligent Caching
Model health data is cached by region with a 60-second TTL to optimize performance:
- Regional Cache - Each region's models_list is cached independently
- Cache Invalidation - Automatic expiration after 60 seconds
- Force Refresh - Click "Refresh" button to bypass cache and fetch fresh data
- Parallel Fetching - All three regions are fetched simultaneously on initial load
This caching strategy ensures fast loading while maintaining near real-time accuracy.
Status Indicators
| Status | Description |
|---|---|
| ✅ Healthy | Model is operating normally (health score ≥ 80) |
| ⚠️ Degraded | Model experiencing some issues (health score 50-79) |
| ❌ Unhealthy | Model is experiencing significant issues (health score < 50) |
| ⚪ Healthy (No data) | No monitoring data available - assumed healthy |
Metrics Explained
- Health Score: Composite score based on error rate and latency (0-100)
- Error Rate: Percentage of failed requests in the last hour
- Avg Latency: Average response time in milliseconds
- Reliability: Long-term reliability metric based on historical EMA
- Total Windows: Number of monitoring intervals collected
Loading model status from all regions...