Status
Every number here comes out of the gateway’s own request log. A model nobody has called says so, instead of showing a percentage built on nothing.
All systems operational
trailing 24 hours- Success rate
- 100.00%
- 0 upstream failures
- Median latency
- 2532 ms
- p50 across all models
- p95 latency
- 3797 ms
- slowest 5% start here
- Requests
- 12
- 7 models with traffic
Last 60 days
14 of 15 days with traffic were clean
2026-07-15today
A pale bar is a day with no recorded traffic, not a day of downtime.
By model
| Model | Status | Success | p50 | p95 | Requests |
|---|---|---|---|---|---|
| GPT-6 Astra400K context | Too few calls | — | 3199 ms | 3199 ms | 2 |
| GPT-5.6 Sol400K context | No traffic | — | — | — | 0 |
| GPT-5.6 Terra400K context | No traffic | — | — | — | 0 |
| GPT-5.5400K context | No traffic | — | — | — | 0 |
| Claude Fable 5200K context | Too few calls | — | — | — | 1 |
| Claude Opus 51M context | No traffic | — | — | — | 0 |
| Claude Sonnet 51M context | No traffic | — | — | — | 0 |
| Gemini 3.8 Flash1M context | Too few calls | — | 2862 ms | 3797 ms | 4 |
| Gemini 3.7 Flash1M context | No traffic | — | — | — | 0 |
| Gemini 3.1 Pro Preview1M context | No traffic | — | — | — | 0 |
| Gemini 3.1 Flash Lite1M context | No traffic | — | — | — | 0 |
| Gemini 3 Flash1M context | Too few calls | — | 1641 ms | 1641 ms | 2 |
| Grok 4.6256K context | No traffic | — | — | — | 0 |
| Grok 4.5256K context | No traffic | — | — | — | 0 |
| MiniMax M3200K context | Too few calls | — | 3406 ms | 3406 ms | 1 |
| MiniMax M2.7 HighSpeed200K context | No traffic | — | — | — | 0 |
| MiniMax M2.7200K context | Too few calls | — | 1367 ms | 1367 ms | 1 |
| MiniMax M2.5 HighSpeed200K context | No traffic | — | — | — | 0 |
| MiniMax M2.5200K context | Too few calls | — | 2185 ms | 2185 ms | 1 |
How this is measured
- A failure means the upstream provider returned 5xx. A 402 over quota or a 429 over your rate limit is the gateway working as designed, and is not counted against it.
- Latency is measured end to end at the gateway, including the upstream call, over successful requests only.
- Below five calls in the window there is no percentage, because at that sample size one bad request reads as a twenty percent failure rate.