> ## Documentation Index
> Fetch the complete documentation index at: https://docs.belvedir.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Hosted Inference

> Trace agents calling hosted OpenAI-compatible endpoints like Together, Fireworks, Groq, and OpenRouter.

Open models too big for your hardware run the same way from a hosted OpenAI-compatible endpoint — the integration is identical to [local models](/inference/local-models), just with a different base URL and a real API key.

## 1. Pick an endpoint

Providers like [Together](https://www.together.ai), [Fireworks](https://fireworks.ai), [Groq](https://groq.com) and [OpenRouter](https://openrouter.ai) serve open models behind OpenAI-compatible APIs. If you'd rather run your own weights, self-host [vLLM](https://docs.vllm.ai) on a GPU cloud — it exposes the same API.

## 2. Point your agent at the hosted endpoint

Same OpenAI SDK, different base URL:

```ts theme={null}
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.together.xyz/v1",
  apiKey: process.env.TOGETHER_API_KEY,
});

const res = await client.chat.completions.create({
  model: "Qwen/Qwen3-235B-A22B",
  messages: [{ role: "user", content: "hello" }],
});
```

## 3. Trace it as usual

Initialize the SDK with `instrumentModules: { openAI: openai.OpenAI }` and wrap agent runs in `withSession`. Spans arrive with the hosted model name, so local, hosted and cloud-provider calls stay distinguishable in the dashboard.

> Note: unlike local models, the hosted provider sees your prompts — the usual data-handling considerations for a cloud provider apply.