From what i can see, Vapi API shows 100% uptime. There are a few latency spikes (especially one large one around ~8pm) Status is marked Degraded, not fully down. That usually means, Higher response times, Some features (LLM calls, transfers, tools, etc.) may be slower or intermittently failing, Core API is still reachable, If your agents are ending calls or behaving oddly, it’s likely related to the degraded performance (especially if you're seeing LLM or tool errors). What you can do right now, Retry failed calls, Avoid pushing production traffic if possible, Monitor the status page for resolution updates, Are you currently seeing call failures, LLM errors, or just latency?