Start typing to search.
Ranks providers by the uptime OpenRouter measured during the latest 30-minute window.
SiliconFlow currently ranks first at 100.00% via siliconflow/fp8.
| Rank | Provider / route | Uptime (last 30 min) | Speed measurements | Endpoint details |
|---|---|---|---|---|
| #1 | Provider / route SiliconFlow siliconflow/fp8 | Uptime (last 30 min)100.00% | Speed measurements
| Endpoint details
|
| #2 | Provider / route Baseten baseten/fp4 | Uptime (last 30 min)100.00% | Speed measurements
| Endpoint details
|
| #3 | Provider / route Amazon Bedrock amazon-bedrock | Uptime (last 30 min)100.00% | Speed measurements
| Endpoint details
|
| #4 | Provider / route Cerebras cerebras/fp16 | Uptime (last 30 min)99.93% | Speed measurements
| Endpoint details
|
| #5 | Provider / route DeepInfra deepinfra/turbo | Uptime (last 30 min)99.91% | Speed measurements
| Endpoint details
|
| #6 | Provider / route OpenInference open-inference/int8 | Uptime (last 30 min)99.81% | Speed measurements
| Endpoint details
|
| #7 | Provider / route Groq groq | Uptime (last 30 min)99.64% | Speed measurements
| Endpoint details
|
| #8 | Provider / route Phala phala | Uptime (last 30 min)99.47% | Speed measurements
| Endpoint details
|
| #9 | Provider / route SambaNova sambanova | Uptime (last 30 min)99.03% | Speed measurements
| Endpoint details
|
| #10 | Provider / route DigitalOcean digitalocean | Uptime (last 30 min)98.54% | Speed measurements
| Endpoint details
|
| #11 | Provider / route Google Vertex google-vertex/global | Uptime (last 30 min)97.58% | Speed measurements
| Endpoint details
|
| #12 | Provider / route Weights & Biases wandb/fp4 | Uptime (last 30 min)97.23% | Speed measurements
| Endpoint details
|
| #13 | Provider / route DekaLLM dekallm/bf16 | Uptime (last 30 min)96.73% | Speed measurements
| Endpoint details
|
| #14 | Provider / route Parasail parasail/fp4 | Uptime (last 30 min)95.94% | Speed measurements
| Endpoint details
|
| #15 | Provider / route MARA mara | Uptime (last 30 min)91.76% | Speed measurements
| Endpoint details
|
| #16 | Provider / route Together together | Uptime (last 30 min)88.78% | Speed measurements
| Endpoint details
|
| #17 | Provider / route Mancer mancer/fp4 | Uptime (last 30 min)88.11% | Speed measurements
| Endpoint details
|
| #18 | Provider / route Nebius nebius/fp4 | Uptime (last 30 min)85.41% | Speed measurements
| Endpoint details
|
| #19 | Provider / route NovitaAI novita/fp4 | Uptime (last 30 min)80.20% | Speed measurements
| Endpoint details
|
| #20 | Provider / route DeepInfra deepinfra/bf16 | Uptime (last 30 min)70.18% | Speed measurements
| Endpoint details
|
| — | Provider / route Amazon Bedrock amazon-bedrock/eu-west-1 | Uptime (last 30 min)Insufficient recent data | Speed measurements
| Endpoint details
|
Endpoint data fetched .
gpt-oss-120b endpoint guide
20 of 21 gpt-oss-120b endpoints currently have the published data required for this ranking. SiliconFlow leads at 100.00% via siliconflow/fp8. The table keeps unranked routes visible so missing measurements do not look like missing provider availability.
The reliability ranking uses the published 30-minute endpoint uptime measurement. This short operational window can reveal current instability, but it is not a service-level agreement or a substitute for longer-term availability monitoring.
Baseten currently ranks second at 100.00%. Compare that gap with input and output price, response-start time, output speed, 30-minute uptime, context length, and the exact route before deciding whether first place is meaningful for your workload.
gpt-oss-120b has 21 published provider options, and performance data is matched to each exact OpenRouter routing tag. Different quantization, context, regional deployment, or provider configuration can change price and behavior even when the underlying model name is identical.
