gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license. It uses a Mixture-of-Experts (MoE) architecture with 3.6B active parameters per forward pass, optimized for...
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% | 100.00% | |
| Operational | 100.00% | 100.00% | 99.86% | |
| Operational | 100.00% | 100.00% | 99.38% | |
| Operational | 100.00% | 100.00% | 100.00% | |
| Operational | 100.00% | 96.15% | 99.60% | |
| Operational | 100.00% | 100.00% | 100.00% | |
| Operational | 100.00% | 83.33% | 86.08% | |
| Operational | — | 100.00% | 100.00% | |
| Operational | 100.00% | 100.00% | 100.00% | |
| Operational | 100.00% | 100.00% | 99.19% | |
| Operational | 100.00% | 100.00% | 99.65% |