gpuos

Integration · Framework

Vercel AI SDK with your own GPUs

Stream gpuos models into Next.js and React apps with the AI SDK's OpenAI-compatible provider.

Base URL
https://gpuos.si/v1
API key
gpuos_key_… from the dashboard
Model
a catalog id, e.g. qwen3-32b

Create a gpuos provider

Install
npm install ai @ai-sdk/openai-compatible
TypeScript
import { createOpenAICompatible } from "@ai-sdk/openai-compatible"
import { streamText } from "ai"

const gpuos = createOpenAICompatible({
  name: "gpuos",
  baseURL: "https://gpuos.si/v1",
  apiKey: process.env.GPUOS_API_KEY,
})

const result = streamText({
  model: gpuos("qwen3-32b"),
  prompt: "Draft a release note for our new dashboard.",
})
for await (const text of result.textStream) process.stdout.write(text)

In a Next.js route handler, return result.toUIMessageStreamResponse() to stream into useChat on the client.

Official documentation: ai-sdk.dev

Questions

Do embeddings work with the AI SDK?
Yes: gpuos.textEmbeddingModel('bge-m3') with embed or embedMany.
Where should the API key live?
In a server-only environment variable. Call gpuos from route handlers or server actions, never from client components.

Related

Use Vercel AI SDK with models on your GPUs

Free for one GPU. Connect a machine, deploy a model, then paste the base URL and your key.

Start freeRead the quickstart