Provider reliability · tracked daily
OpenAI: gpt-oss-120b (batch): which provider is most reliable?
Across 23 providers, Groq is the most reliable way to run OpenAI: gpt-oss-120b (batch), at 100.0% uptime and 125ms median latency. Measured directly from OpenRouter's endpoint data, tracked daily.
- Groq - 100.0% uptime · 125ms
- BaseTen - 100.0% uptime · 148ms
- Cerebras - 100.0% uptime · 154ms
- Crusoe - 100.0% uptime · 157ms
- CoreWeave - 100.0% uptime · 264ms
- Nebius - 100.0% uptime · 266ms
- Together - 100.0% uptime · 295ms
- AkashML - 100.0% uptime · 307ms
- DeepInfra - 100.0% uptime · 336ms
- Parasail - 100.0% uptime · 370ms
Uptime is OpenRouter's own trailing-30-minute metric, captured once at our last daily sync - a snapshot of that window, not a live/real-time feed.
| # | Provider | Uptime (30m) | Latency p50 | Throughput |
|---|---|---|---|---|
| 01 | Groq | 100.0% | 125ms | 371 tok/s |
| 02 | BaseTen | 100.0% | 148ms | 240 tok/s |
| 03 | Cerebras | 100.0% | 154ms | 813 tok/s |
| 04 | Crusoe | 100.0% | 157ms | 245 tok/s |
| 05 | CoreWeave | 100.0% | 264ms | 87 tok/s |
| 06 | Nebius | 100.0% | 266ms | 302 tok/s |
| 07 | Together | 100.0% | 295ms | 75 tok/s |
| 08 | AkashML | 100.0% | 307ms | 131 tok/s |
| 09 | DeepInfra | 100.0% | 336ms | 46 tok/s |
| 10 | Parasail | 100.0% | 370ms | 131 tok/s |
| 11 | Amazon Bedrock | 100.0% | 432ms | 202 tok/s |
| 12 | DekaLLM | 100.0% | 538ms | 78 tok/s |
| 13 | DigitalOcean | 100.0% | 633ms | 67 tok/s |
| 14 | Novita | 100.0% | 752ms | 144 tok/s |
| 15 | Near AI | 100.0% | 806ms | 41 tok/s |
| 16 | Mancer 2 | 100.0% | 914ms | 47 tok/s |
| 17 | Mara | 100.0% | 1086ms | 186 tok/s |
| 18 | SiliconFlow | 100.0% | 1330ms | 35 tok/s |
| 19 | Venice | 100.0% | 1483ms | 43 tok/s |
| 20 | 0.0% | 266ms | 98 tok/s | |
| 21 | SambaNova | - | 826ms | 280 tok/s |
| 22 | Phala | - | 1144ms | 123 tok/s |
| 23 | Sail Research | - | - | - |
Uptime over time
- AkashML
- Amazon Bedrock
- BaseTen
- Cerebras
- CoreWeave
- Crusoe
Frequently asked questions
Which provider is most reliable for OpenAI: gpt-oss-120b (batch)?
Groq is the most reliable provider for OpenAI: gpt-oss-120b (batch), at 100.0% uptime (OpenRouter's trailing-30-minute metric, snapshotted at our last daily sync) and 125ms median latency, out of 23 providers tracked.
What is OpenAI: gpt-oss-120b (batch)'s uptime?
The best-performing provider for OpenAI: gpt-oss-120b (batch) reports 100.0% uptime, tracked daily.