---
title: AI
description: A thin wrapper over the AI SDK, with chat primitives.
type: reference
related:
  - /en/docs/examples/ai-chatbot
---

# AI



`@kreogen/ai` re-exports the [AI SDK](https://ai-sdk.dev) and adds two things:
a configured model registry, and the minimum UI needed to render a
conversation.

It is deliberately thin. There is no chat abstraction of kreogen's own to learn
or work around — everything the SDK documents works here unchanged.

## Entry points

| Import                           | Provides                                                                             |
| -------------------------------- | ------------------------------------------------------------------------------------ |
| `@kreogen/ai`                    | Everything from `ai` — `streamText`, `generateObject`, `DefaultChatTransport`, types |
| `@kreogen/ai/lib/react`          | Everything from `@ai-sdk/react` — `useChat`, `useCompletion`                         |
| `@kreogen/ai/lib/models`         | The configured `models` object                                                       |
| `@kreogen/ai/components/thread`  | `<Thread>`, a scrolling conversation container                                       |
| `@kreogen/ai/components/message` | `<Message>`, one bubble, markdown-aware                                              |

<Callout type="warn">
  This package is on AI SDK **v6** (`ai@^6`, `@ai-sdk/react@^3`). Guides written
  for v4 will not work: `useChat` no longer owns the input, and
  `toDataStreamResponse()` is `toUIMessageStreamResponse()`. The
  [chatbot example](/en/docs/examples/ai-chatbot#what-changed-since-v4) has the
  full mapping.
</Callout>

## Models

```ts title="packages/ai/lib/models.ts"
const { OPENAI_API_KEY } = keys();

const openai = OPENAI_API_KEY
  ? createOpenAI({ apiKey: OPENAI_API_KEY })
  : undefined;

export const models:
  | {
      chat: LanguageModel;
      embeddings: EmbeddingModel;
    }
  | undefined = openai
  ? {
      chat: openai('gpt-4o-mini'),
      embeddings: openai.textEmbeddingModel('text-embedding-3-small'),
    }
  : undefined;
```

Naming the models in one place is what lets you change provider or tier in one
edit rather than across every call site. Swap `@ai-sdk/openai` for another
provider package and the rest of the repository is unaffected — the SDK's model
interface is the boundary.

The type annotation is not decorative. Without it TypeScript infers a type
naming the hoisted `@ai-sdk/provider` path, which is not portable across
installs and breaks in a way that has nothing obviously to do with models.

`models` is `undefined` without `OPENAI_API_KEY`, like every other optional
integration here. The guard is not defensive tidiness: the provider constructs
happily without a key — it only reads one per request — so before it existed
the first model call failed with a `LoadAPIKeyError` at runtime while the type
insisted `chat` was always there, leaving a caller no signal that it had to
degrade.

## Configuration

| Variable         | Required | Purpose               |
| ---------------- | -------- | --------------------- |
| `OPENAI_API_KEY` | no       | Must start with `sk-` |

Without the key, `models` is `undefined` and every AI feature has to say so.
Guard with `models?.` at the seam, or answer 503 from the route handler — see
the [chatbot example](/en/docs/examples/ai-chatbot).

## Components

`<Thread>` is a flex column that scrolls. `<Message>` takes a `UIMessage` and
renders it, aligned by role, with the assistant's markdown streamed through
[Streamdown](https://streamdown.ai) so partial output does not flicker as
syntax completes.

```tsx
<Thread>
  {messages.map((message) => (
    <Message data={message} key={message.id} />
  ))}
</Thread>
```

`<Message>` reads `message.parts`, filtering to the text ones. In v6 a message
is an array of typed parts — text, reasoning, tool calls, files — and
`message.content` no longer exists. A component reading it renders nothing
rather than failing, which is a slow bug to find.

## Things worth doing per project

**Authenticate the route.** A handler that calls a paid model on request is
somebody else's bill if it is public. Route handlers are not covered by the
page-level auth check.

**Rate limit it.** [A limiter](/en/docs/packages/rate-limit) keyed on the user
id bounds the damage from a stuck client as much as from an attacker.

**Bound the tool loop.** `stopWhen: stepCountIs(n)` on `streamText` caps how
many round trips one request may make. Without it, a model that keeps calling
tools keeps spending.


---

For a semantic overview of all documentation, see [/sitemap.md](/sitemap.md)

For an index of all available documentation, see [/llms.txt](/llms.txt)