Compared with GLM-4.5, this generation brings several key improvements: Longer context window: The context window has been expanded from 128K to 200K tokens, enabling the model to handle more complex...
Share of successful requests per carrier, measured over the last 5 minutes, 30 minutes and day. When one carrier dips, the router fails over to the next.
| Provider | Status | Last 5m | Last 30m | Last day |
|---|---|---|---|---|
| Operational | 100.00% | 100.00% | 99.79% | |
| Operational | 100.00% | 100.00% | 99.89% | |
| Operational | 100.00% | 100.00% | 100.00% | |
| Operational | — | — | 100.00% |