Use Flux with the Vercel AI SDK

The Vercel AI SDK talks to FluxRouter through its OpenAI-compatible provider. You create the provider with baseURL set to https://api.fluxrouter.ai/v1 and your Flux key, then use flux-auto as the model.

Set up the OpenAI-compatible provider#

Install the provider package (npm install @ai-sdk/openai-compatible ai), then create a Flux provider:

import { createOpenAICompatible } from "@ai-sdk/openai-compatible";
import { generateText } from "ai";

const flux = createOpenAICompatible({
  name: "flux",
  baseURL: "https://api.fluxrouter.ai/v1",   // the one line that points at Flux
  apiKey: process.env.FLUX_API_KEY,          // your Flux key
});

const { text } = await generateText({
  model: flux("flux-auto"),                  // let Flux route, or pin a flux-* id
  prompt: "Say hello from FluxRouter.",
});

console.log(text);

createOpenAICompatible is the AI SDK's documented way to point at any OpenAI-compatible endpoint.

Stream a response#

Streaming works the same way; swap generateText for streamText:

import { streamText } from "ai";

const result = streamText({
  model: flux("flux-auto"),
  prompt: "Stream a short greeting from FluxRouter.",
});

for await (const chunk of result.textStream) {
  process.stdout.write(chunk);
}

Pick a model#

flux-auto routes each request for you. To pin a model, pass a tier alias (flux-fast, flux-standard, flux-reasoning) or a flux-pinned-* id to the provider, for example flux("flux-reasoning"). See Models.