# API Reference (/desktop/api-preference)

Jan's local API server exposes an OpenAI-compatible REST API at `http://127.0.0.1:1337`. Use it as a drop-in replacement for cloud APIs in any application that supports OpenAI-compatible endpoints.

## Supported Endpoints [#supported-endpoints]

### `GET /v1/models` [#get-v1models]

Returns a list of all models currently loaded or available in Jan.

```bash
curl http://127.0.0.1:1337/v1/models \
  -H "Authorization: Bearer YOUR_API_KEY"
```

```json
{
  "object": "list",
  "data": [
    { "id": "jan-v3-4b-base-instruct", "object": "model" }
  ]
}
```

---

### `POST /v1/chat/completions` [#post-v1chatcompletions]

OpenAI-compatible chat completions endpoint. Supports streaming, tool calling, and multi-turn conversations.

```bash
curl http://127.0.0.1:1337/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "jan-v3-4b-base-instruct",
    "messages": [{"role": "user", "content": "Hello!"}],
    "stream": false
  }'
```

---

### `POST /v1/messages` [#post-v1messages]

Anthropic-compatible messages endpoint. Jan automatically translates requests to the internal format, so you can use Anthropic SDK clients pointed at your local server.

```bash
curl http://127.0.0.1:1337/v1/messages \
  -H "Content-Type: application/json" \
  -H "x-api-key: YOUR_API_KEY" \
  -d '{
    "model": "jan-v3-4b-base-instruct",
    "max_tokens": 1024,
    "messages": [{"role": "user", "content": "Hello!"}]
  }'
```

---

<Callout type="info">
**OpenAI Responses API** (`/v1/responses`) support is coming soon.
</Callout>

## Use With Agents & Integrations [#use-with-agents--integrations]

Jan's local API is designed to wire directly into AI agents and coding tools — no cloud account needed.

<DocCards cols={2}>
  <DocCard title="Claude Code" href="/desktop/integrations/claude-code" icon={<Code2 size={20} />}>Point Claude Code at Jan's local server to run Opus, Sonnet, and Haiku tiers on your own hardware.</DocCard>
  <DocCard title="OpenClaw" href="/desktop/integrations/openclaw" icon={<Bot size={20} />}>Wire OpenClaw to Jan's local API for a fully private autonomous agent — no cloud, no usage fees.</DocCard>
</DocCards>

## Using with OpenAI SDK [#using-with-openai-sdk]

Point the OpenAI SDK at your local server by setting the base URL and API key:

```python
from openai import OpenAI

client = OpenAI(
    base_url="http://127.0.0.1:1337/v1",
    api_key="YOUR_API_KEY"
)

response = client.chat.completions.create(
    model="jan-v3-4b-base-instruct",
    messages=[{"role": "user", "content": "Hello!"}]
)
```

```typescript
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "http://127.0.0.1:1337/v1",
  apiKey: "YOUR_API_KEY",
});
```