Real-time uptime and latency monitoring for 11 major AI API providers. Your agent checks once, routes smart, and never wastes tokens on a dead endpoint.
Every time your agent sends a request to a down API, you burn tokens on the retry. STATUS probes all 11 major AI providers every 60 seconds and caches the results. Your agent checks one endpoint, gets a clean JSON response, and routes accordingly.
Response includes p50/p95 latency per provider. Route to the fastest available model, not just any available one.
When your primary provider is down, STATUS tells you which alternatives are up — in a single request, not 11 sequential ones.
Paid tier includes 7-day uptime percentage per provider. Make data-driven routing decisions, not guesses.
Clean JSON output, MCP-compatible, no HTML parsing. One request, all providers, structured data ready for your context window.
We probe every provider's chat completions endpoint every 60 seconds from our VPS in Nuremberg. Free tier gives you current status + latency. Paid tier adds uptime history.
Get current status and latency for all 11 providers.
curl "https://status.4bots.ai/v1/status"
Example response:
{
"providers": {
"openai": { "status": "up", "latency_ms": 142 },
"anthropic": { "status": "up", "latency_ms": 218 },
"groq": { "status": "up", "latency_ms": 45 },
"mistral": { "status": "degraded", "latency_ms": 891 },
"deepseek": { "status": "down", "latency_ms": null },
...
},
"cached_at": "2026-06-17T14:23:00Z",
"free_remaining": 847
}
Check a single provider. Useful when you only route between two.
curl "https://status.4bots.ai/v1/status/anthropic"
Health check. Always free, always returns 200.
curl "https://status.4bots.ai/health"
STATUS is our loss-leader. The free tier is generous enough for most agent workloads. Paid tier unlocks higher rate limits and historical data.
No signup. No API key. Just send a request.
How x402 works