What makes a good fallback model?
One that covers the same required capabilities and stays healthy enough under load. The checker confirms both before you wire it into a routing policy.
/ tools / uk / llm-routing-health-checker
LLM Routing Health Checker
Check primary and fallback fit from health and capabilities.
Check whether a model fits as a primary or fallback route from its health, success rate, latency, and required capabilities — so your routing policy degrades gracefully instead of failing under load.
- Test primary and fallback fit against required capabilities
- Factor in success rate and p50/p95 latency
- Catch capability gaps before they hit production
Routing Health
Inputs are processed in this browser; do not paste live API keys into this diagnostic view.
Diagnostic report
- primaryFit
- true
- fallbackFit
- true
- missingCapabilities
- []
- latencyRange
- [2300,6800]
- successRate
- 0.99
- warnings
- []
FAQ
LLM Routing Health Checker FAQ
How does model routing improve reliability?
Routing sends traffic to a healthy primary and shifts to a fallback when success rate or latency degrades, avoiding single-provider outages.
What if health data is unavailable?
The checker still evaluates capability fit and flags that a live health probe is missing so you can treat the route as fallback-only.
Related tools
Keep optimising cost
Next step
Use the report as the next integration decision.
Copy the base URL, compare model details, or create a key when you are ready to run a real compatibility test.