Inference provider

Use Fireworks AI through Relays.

Route Fireworks AI through Relays’ maintained OpenAI-compatible Chat Completions preset.

One alias in front of Fireworks AI.

Your application calls a Relays alias. Relays holds the Fireworks AI credential, applies the route policy, and hands back the provider response in the shape your SDK expects.

  1. CallerYour application

    Keeps the OpenAI-compatible request it already sends.

  2. EdgeRelays

    Holds the key, picks the candidate, writes the receipt.

  3. UpstreamFireworks AI

    Receives the call on its own base URL.

Wire protocol
OpenAI-compatible
Relays endpoints
  • Chat Completions
Credential boundary
Bring your own Fireworks API keyFIREWORKS_API_KEY
Upstream base URL
https://api.fireworks.ai/inference/v1
Connection
Maintained Relays preset
Source
Fireworks AI API documentation

Relays’ Fireworks AI preset keeps the provider key server-side and exposes the OpenAI-compatible Chat Completions contract through a public Relays alias.

What is measured, and what is not.

Availability comes from Fireworks AI. Latency, error rate, and cost per successful request come from your own traffic, so Relays leaves them empty until requests run.

Upstream status

Fireworks AI
Not published to Relays
Fireworks AI does not publish a status feed Relays can read, so no availability is reported here.

Relays measurements

Your traffic
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.

Models at the edge.

Model IDs sit behind a Relays alias, so a catalogue change at Fireworks AI never reaches your code.

Catalogue not connected
Relays has not published a Fireworks AI model list on this page. Route candidates are still set by model ID in your route configuration.

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.

Fallback

Candidates stay inside the contract.

Fireworks AI can be an explicit candidate or fallback in an eligible Chat Completions route.

See how a route is chosen
Budget

A ceiling on this route.

Set an alert threshold and a stop threshold for the route rather than for the whole application. Relays checks both before the call goes upstream.

Receipt

Each attempt is written down.

Every call records the alias, the candidates tried, the status each returned, tokens in and out, and the estimated cost of the attempt that served it.

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 Fireworks AI route with you.

Start a conversation