Latency and status
Measured from our own health checks, per host per model. Public, no sign-in.
Some systems need attentionWaiting for the first check
API (chat, messages, responses)
Outage
Dashboard
Operational
Billing and top-ups
Not responding
Model catalog
Operational
Median time to first token, all pinned hosts, hourly
No health checks in the last 24 hours.
Per host, per model
p50 over the last 24 hours| Model | Host | Time to first token | Throughput | Uptime | Last 30 days |
|---|---|---|---|---|---|
| DeepSeek V4.1 Flashdeepseek/deepseek-v4.1-flash | DeepSeek, no recent checks pinned | - | - | - | |
| Step 5 Previewstepfun/step-5-preview | StepFun, no recent checks pinned | - | - | - | |
| GLM 5.3 Flashz-ai/glm-5.3-flash | Z.AI, no recent checks pinned | - | - | - | |
| MiMo-V2.6-Flashxiaomi/mimo-v2.6-flash | Xiaomi, no recent checks pinned | - | - | - | |
| Hy4 previewtencent/hy4-preview | Tencent, no recent checks pinned | - | - | - | |
| Nemotron 3 Ultra (free)nvidia/nemotron-3-ultra-550b-a55b:free | Nvidia, no recent checks pinned | - | - | - | |
| GPT-6 Lunaopenai/gpt-6-luna | OpenAI, no recent checks pinned | - | - | - | |
| Claude Opus 5.5anthropic/claude-opus-5.5 | Anthropic, no recent checks pinned | - | - | - | |
| DeepSeek V4 Flash 0731deepseek/deepseek-v4-flash-0731 | Reka, no recent checks pinned | - | - | - | |
| GLM 5.3z-ai/glm-5.3 | Z.AI, no recent checks pinned | - | - | - |
Recent incidents
No incidents recorded.
A grey bar is a day with no checks. Each read covers the newest 1,000 checks in the range.