Status
Availability of the gateway and the upstream model API.
Some models are slower than usual
36 of 37 models online
Served by: Primary route
🟢 Live🟡 Slow🔴 Offline
Gemini
1LiveModelSuccessLatencyInput / 1M
GPT
8LiveModelSuccessLatencyInput / 1M
Claude
8SlowModelSuccessLatencyInput / 1M
Grok
3LiveModelSuccessLatencyInput / 1M
DeepSeek
8LiveModelSuccessLatencyInput / 1M
- deepseek-v4.1-flash100%3.2s$0.04/M
- deepseek-v4.1-flash-101790%2.3s$0.04/M
- deepseek-v4.1-flash-1017-abliterated100%2.9s$0.30/M
- deepseek-v4.1-flash-abliterated100%3.8s$0.30/M
- deepseek-v4.1-flash-abliterated-fast90%2.1s$1.00/M
- deepseek-v4-flash-0731100%2.9s$0.06/M
- deepseek-v4-pro-0813100%4.1s$0.50/M
- deepseek-v4-pro-uncensored100%5.9s$0.50/M
GLM
5LiveModelSuccessLatencyInput / 1M
MiMo
2LiveModelSuccessLatencyInput / 1M
Kimi
2LiveModelSuccessLatencyInput / 1M
Every model is tested with a real one-token request. Results refresh continuously.