System status
All systems operational.
Real-time health for the Nimbus gateway and every upstream provider we route to. Uptime measured continuously since 2026-01-14.
Overall uptime (30d)
99.94%
Gateway p50
42ms
Providers monitored
12
Open incidents
0
Provider health
Refreshed every 60 secondsNimbus Gateway
All regions healthy.
p50
42ms
Azure OpenAI (East US 2)
Committed capacity nominal.
p50
380ms
Azure OpenAI (Sweden Central)
Secondary pool, healthy.
p50
410ms
Anthropic (via SpiderSense)
Claude Opus and Sonnet families online.
p50
520ms
Google Vertex AI
Gemini 2.5 Pro seeing intermittent 5xx from upstream — retrying via secondary region.
p50
640ms
xAI Grok
Grok-4 and Grok-4-fast serving normally.
p50
590ms
DeepSeek
DeepSeek-V3 and DeepSeek-R1 available.
p50
720ms
Mistral La Plateforme
Codestral and Mistral Large healthy.
p50
460ms
Meta (via Groq)
Llama-3.3-70B and Llama-4 available.
p50
210ms
Cohere
Command R+ healthy.
p50
480ms
Dashboard + Website
Vercel edge network nominal.
p50
88ms
Billing (SellAuth + NowPayments)
Fiat and crypto rails both accepting payments.
p50
1100ms
Recent incidents
minor2026-06-28Duration: 34 min
Elevated latency on Gemini 2.5 Pro (us-central1)
Upstream Vertex region degraded. Automatically rerouted to us-east4 pool. No requests dropped.
minor2026-06-19Duration: 12 min
Dashboard slow to load
Vercel edge cache invalidation storm during a deploy. Rolled back the deploy; latency returned to baseline.
major2026-06-04Duration: 1 h 47 min
Anthropic upstream rate-limit event
Anthropic imposed short-term throughput caps affecting our shared pool. Traffic auto-rerouted to Azure OpenAI equivalents where models overlapped. Postmortem posted 2026-06-06.
minor2026-05-22Duration: 8 min
Gateway 502 spike
Bad edge worker deploy in one region. Rollback completed inside SLA. Affected roughly 0.3% of requests globally during the window.