Status
All systems operational
Live from real gateway traffic at 5-minute resolution — no synthetic probes. Quiet windows are assumed operational. Last 30 days.
Hosted models
glm-4.7-flashxoperational—qwen3.5-flashoperational—glm-5.3-flashoperational—deepseek-v4-flashoperational—minimax-m3operational—qwen3-coder-nextoperational—qwen3.7-plusoperational—qwen3.8-27boperational—kimi-k2.7-codeoperational—deepseek-v4-prooperational—glm-5.2operational—glm-5.3operational—qwen3.8-2.4t-a95boperational—qwen3.8-maxoperational—kimi-k3operational—uptime 30d · average time-to-first-token, last 24h
Product
0requests, 24h
4requests, 30d
0upstream failures, 30d
20Ktokens in, 30d
2092tokens out, 30d
51.7%cache hit-rate, 30d
858Ktokens in, all time
67Ktokens out, all time
Past incidents
No incidents in the last 30 days.
Machine-readable: GET /status.json · build e2ff9bccfbb633e565a4fe34980d134b37b62e2b4fff2fcccbabce4421446887 · generated 2026-10-17 UTC