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

# Quick Start

> Your first ```/chat/completions``` call

Hebo Gateway provides a unified API for completions, embeddings, and model discovery.
It normalizes provider differences so you can switch models without rewriting your client.

## Why Gateway

* OpenAI-compatible endpoints and normalized model IDs across providers
* One way to configure reasoning across all models
* Support for streaming and non-streaming responses, as well as reasoning and tool calls
* Compatible with common AI SDKs (Vercel AI SDK, TanStack AI, Langchain, OpenAI SDK, ...)

## Choose an SDK

If you havent installed any SDK yet, you'll have to do that first.

Most of our own examples in the documentation will refer to Vercel AI SDK (which we use ourselves internally) using TypeScript.

<CodeGroup>
  ```ts Vercel AI SDK theme={null}
  bun add ai @ai-sdk/openai-compatible
  ```

  ```ts OpenAI SDK theme={null}
  bun add openai
  ```

  ```ts LangChain theme={null}
  bun add @langchain/openai
  ```
</CodeGroup>

<Info>
  We recommend `bun` as your default JavaScript / TypeScript toolkit, but you can also use
  `npm`, `pnpm` or `yarn`.
</Info>

You can use any other SDKs and languages (e.g. Python), as long as they can connect to an OpenAI-compatible endpoint.

If you find any issues accessing Hebo Gateway with another library, we would appreciate a report on our [issue tracker](https://github.com/heboai/hebo/issues).

## Your First Call

Now you're already ready for your first call:

<CodeGroup>
  ```ts Vercel AI SDK theme={null}
  import { createOpenAICompatible } from "@ai-sdk/openai-compatible";
  import { generateText } from "ai";

  const hebo = createOpenAICompatible({
    name: "hebo",
    apiKey: process.env.HEBO_API_KEY,
    baseURL: "https://gateway.hebo.ai/v1",
  });

  const { text } = await generateText({
    model: hebo("openai/gpt-oss-20b"),
    prompt: "Tell me a joke about monkeys",
  });

  console.log(text);
  ```

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

  const client = new OpenAI({
    apiKey: process.env.HEBO_API_KEY,
    baseURL: "https://gateway.hebo.ai/v1",
  });

  const response = await client.chat.completions.create({
    model: "openai/gpt-oss-20b",
    messages: [{ role: "user", content: "Tell me a joke about monkeys" }],
  });

  console.log(response.choices[0]?.message?.content);
  ```

  ```ts LangChain theme={null}
  import { ChatOpenAI } from "@langchain/openai";

  const model = new ChatOpenAI({
    apiKey: process.env.HEBO_API_KEY,
    configuration: { baseURL: "https://gateway.hebo.ai/v1" },
    model: "openai/gpt-oss-20b",
  });

  const response = await model.invoke([{ role: "user", content: "Tell me a joke about monkeys" }]);

  console.log(response.content);
  ```

  ```bash Bash theme={null}
  curl https://gateway.hebo.ai/v1/chat/completions \
    -H "Authorization: Bearer $HEBO_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{
      "model": "openai/gpt-oss-20b",
      "messages": [
        { "role": "user", "content": "Tell me a joke about monkeys" }
      ]
    }'
  ```
</CodeGroup>

<Info>
  If you don't have an API key yet, register for free in the [Hebo
  Console](https://console.hebo.ai).
</Info>

## What's Next?

* Dig deeper into advanced features like [Tool Calling](/gateway/tool-calling), [Structured Outputs](/gateway/structured-outputs), [Reasoning](/gateway/reasoning), and [Embeddings](/gateway/embeddings).