One alias in
front of vLLM.
Your application calls a Relays alias. Relays owns the connection to your vLLM endpoint, applies the route policy, and hands back the provider response in the shape your SDK expects.
- Platform
- OpenAI-compatible
- Lines
- Chat Completions
- Credential
- No upstream key
- Setup
- Built-in preset
- Base URL
- http://localhost:8000/v1
The route.
Relays’ vLLM preset targets a local OpenAI-compatible server with no upstream authentication by default. Use an explicit registry entry when the endpoint or authentication differs.
- Your applicationSends the OpenAI-compatible request it already sends.
- RelaysHolds the key, picks the candidate, writes the receipt.
- vLLMReceives the call on its own base URL.
Who can cover
for vLLM.
vLLM can be part of an explicit Chat Completions route alongside hosted candidates. Route priority makes a local-first policy visible and reviewable.
OpenAIChat CompletionsOpenAI- GeminiChat CompletionsOpenAI-compatible
- Mistral AIChat CompletionsOpenAI-compatible
- DeepSeekChat CompletionsOpenAI-compatible
GroqChat CompletionsOpenAI-compatible
Together AIChat CompletionsOpenAI-compatible
Fireworks AIChat CompletionsOpenAI-compatible- OpenRouterChat CompletionsOpenAI-compatible
- OllamaChat CompletionsOpenAI-compatible
Models at the edge.
Model IDs sit behind a Relays alias, so a catalogue change at vLLM never reaches your code.
No vLLM catalogue is published on this page. Route candidates are still set by model ID in your route configuration.
Measured,
and not yet.
Availability comes from vLLM. Latency, error rate, and cost per successful request come from your own traffic, so Relays leaves them empty until requests run.
- Upstream status
No public feed
vLLM does not publish a status feed Relays can read, so no availability is reported here.
- 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 vLLM route with you.
