One alias in
front of OpenAI.
Your application calls a Relays alias. Relays holds the OpenAI credential, applies the route policy, and hands back the provider response in the shape your SDK expects.
OpenAIOperational- Platform
- OpenAI
- Lines
- Chat Completions · Responses
- Credential
- OPENAI_API_KEY
- Setup
- Registry route
- Base URL
- https://api.openai.com/v1
The route.
Relays keeps the OpenAI request family intact and forces Responses store:false for its stateless gateway boundary.
- Your applicationSends the OpenAI request it already sends.
- RelaysHolds the key, picks the candidate, writes the receipt.
- OpenAIReceives the call on its own base URL.
Who can cover
for OpenAI.
Chat Completions and Responses routes can use ordered fallback candidates that declare the same Relays edge endpoint. Relays does not translate Responses into another API family to make a fallback fit.
- GeminiChat CompletionsOpenAI-compatible
- Mistral AIChat CompletionsOpenAI-compatible
- DeepSeekChat CompletionsOpenAI-compatible
GroqChat Completions · ResponsesOpenAI-compatible
Together AIChat CompletionsOpenAI-compatible
Fireworks AIChat CompletionsOpenAI-compatible- OpenRouterChat CompletionsOpenAI-compatible
- vLLMChat CompletionsOpenAI-compatible
- OllamaChat CompletionsOpenAI-compatible
Models at the edge.
Relays keeps a copy of the OpenAI catalogue and the edge each model is routed to.
- gpt-5.6-solComplex reasoning and codingChat Completions · Responses
- gpt-5.6-terraBalanced intelligence and costChat Completions · Responses
- gpt-5.6-lunaHigh-volume, cost-sensitive workChat Completions · Responses
- gpt-image-2Image generation and editingNot routed
- gpt-realtime-2.1Speech and tool useNot routed
- gpt-realtime-2.1-miniLower-cost realtimeNot routed
- gpt-realtime-2Realtime speech and tool useNot routed
- gpt-realtime-translateSpeech-to-speech translationNot routed
- gpt-realtime-1.5Audio in, audio outNot routed
- gpt-realtime-miniCost-efficient realtime · deprecating at the providerNot declared at Relays edge
- gpt-4o-mini-ttsText-to-speechNot routed
- gpt-transcribeHigh-accuracy speech-to-textNot routed
- gpt-live-transcribeLow-latency live transcriptionNot routed
- gpt-realtime-whisperStreaming speech-to-textNot routed
- gpt-4o-transcribeGPT-4o speech-to-textNot routed
- gpt-4o-mini-transcribeGPT-4o mini speech-to-textNot routed
Catalogue read from OpenAI model catalog on Aug 7, 2026.
Measured,
and not yet.
Availability comes from OpenAI. Latency, error rate, and cost per successful request come from your own traffic, so Relays leaves them empty until requests run.
- Upstream status
Operational
- Chat CompletionsOperational
- ResponsesOperational
Checked Nov 11, 2026, 2:42 PM UTC.
OpenAI status page- Relays measurements
No sample yet
Time to first token, error rate, and cost per successful request are read from requests that ran through Relays. This route has none, so there is nothing to plot.
Ask for an API key
Every route carries a policy.
A route is a list of candidates, a budget, and a record of what happened. You set all three before any traffic runs.
- Candidates
- Fallbacks stay inside the contract: only connections on a shared line can take the call. How a route is chosen
- Ceiling
- An alert threshold and a stop threshold for this route, checked before the call goes upstream.
- Receipt
- Every call records the alias, the candidates tried, each status, tokens in and out, and the estimated cost. Read a route receipt
One endpoint is enough to start.
Keep the client you already wrote. Bring us the workload and we will map the first OpenAI route with you.
