> ## Documentation Index
> Fetch the complete documentation index at: https://docs.x402layer.cc/llms.txt
> Use this file to discover all available pages before exploring further.

# API (OpenAI-compatible)

> Call SGL Grid with the OpenAI Chat Completions API — change the base URL and key, keep your existing SDK.

SGL Grid implements the **OpenAI Chat Completions API**. If your code already uses the OpenAI SDK, point it at the Grid and it works.

**Base URL:** `https://grid.x402compute.cc/v1`

## Quickstart (cURL)

```bash theme={null}
curl https://grid.x402compute.cc/v1/chat/completions \
  -H "X-API-Key: <YOUR_KEY>" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "<model-id>",
    "messages": [{"role": "user", "content": "Explain confidential compute in one line."}],
    "max_tokens": 256
  }'
```

## Drop-in with the OpenAI SDK

```python theme={null}
from openai import OpenAI

client = OpenAI(
    base_url="https://grid.x402compute.cc/v1",
    api_key="<YOUR_KEY>",           # an SGL Grid API key
)

resp = client.chat.completions.create(
    model="<model-id>",
    messages=[{"role": "user", "content": "Hello from the Grid"}],
)
print(resp.choices[0].message.content)
```

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

const client = new OpenAI({
  baseURL: "https://grid.x402compute.cc/v1",
  apiKey: process.env.SGL_API_KEY,
});

const r = await client.chat.completions.create({
  model: "<model-id>",
  messages: [{ role: "user", content: "Hello from the Grid" }],
});
console.log(r.choices[0].message.content);
```

## Authentication

Pass one of:

* `X-API-Key: <key>` — scoped API key (recommended for servers/agents)
* a session bearer token (dashboard)
* a per-request wallet signature

See [Wallet & auth](/cloud/concepts/wallet-auth).

## Paying

* **API key / credits** — billed to the key's wallet at exact token cost.
* **x402** — sign a USDC payment per request (Solana or Base). See [Billing](/cloud/concepts/billing).

## Notes

* Request/response shapes match OpenAI Chat Completions (`model`, `messages`, `max_tokens`, `temperature`, `usage`).
* For maximum privacy (browser-side end-to-end encryption), use the dashboard/SDK confidential path — see [Confidential inference](/cloud/grid/confidential-inference).
* **Streaming** (`stream: true`) is on the roadmap; today set `stream: false`.

<Note>List available models with `GET https://grid.x402compute.cc/grid/models` — see [Models](/cloud/grid/models).</Note>