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

# Streaming Responses

> Real-time incremental token streams using Server-Sent Events (SSE) protocol.

Streaming on the `/v1/chat/completions` endpoint enables your applications to display text tokens as soon as they are produced by the AI model, minimizing perceived latency for chat and terminal interfaces.

<CodeGroup>
  ```python Python (OpenAI SDK) icon="python" theme={"theme":{"light":"ayu-dark","dark":"catppuccin-latte"}}
  from openai import OpenAI
  import os

  client = OpenAI(
      base_url="https://api.neosantara.xyz/v1",
      api_key=os.environ["NEOSANTARA_API_KEY"]
  )

  stream = client.chat.completions.create(
      model="deepseek-v4.1-flash",
      messages=[{"role": "user", "content": "Outline a microservices architecture guide."}],
      stream=True
  )

  for chunk in stream:
      delta = chunk.choices[0].delta.content or ""
      print(delta, end="", flush=True)
  print()
  ```

  ```javascript TypeScript (Node.js) icon="js" theme={"theme":{"light":"ayu-dark","dark":"catppuccin-latte"}}
  import OpenAI from "openai";

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

  const stream = await client.chat.completions.create({
    model: "deepseek-v4.1-flash",
    messages: [{ role: "user", content: "Explain SOLID principles concisely." }],
    stream: true,
  });

  for await (const chunk of stream) {
    process.stdout.write(chunk.choices[0]?.delta?.content || "");
  }
  console.log();
  ```

  ```bash cURL icon="terminal" theme={"theme":{"light":"ayu-dark","dark":"catppuccin-latte"}}
  curl -N -X POST https://api.neosantara.xyz/v1/chat/completions \
    -H "Authorization: Bearer $NEOSANTARA_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{
      "model": "deepseek-v4.1-flash",
      "messages": [{"role": "user", "content": "Hello"}],
      "stream": true
    }'
  ```
</CodeGroup>

## Event Stream Format

Each chunk arrives as an SSE event in `data: <JSON>` format:

```json theme={"theme":{"light":"ayu-dark","dark":"catppuccin-latte"}}
data: {"id":"chatcmpl-abc","object":"chat.completion.chunk","created":1741764000,"model":"deepseek-v4.1-flash","choices":[{"index":0,"delta":{"content":"Hello"},"finish_reason":null}]}
```

When generation finishes, the gateway transmits the standard completion signal:

```text theme={"theme":{"light":"ayu-dark","dark":"catppuccin-latte"}}
data: [DONE]
```

## Tracking Token Usage During Streams

Pass `stream_options` to receive the final token accounting block in the trailing chunk:

```python theme={"theme":{"light":"ayu-dark","dark":"catppuccin-latte"}}
stream = client.chat.completions.create(
    model="deepseek-v4.1-flash",
    messages=[{"role": "user", "content": "Hello"}],
    stream=True,
    stream_options={"include_usage": True}
)

for chunk in stream:
    if chunk.usage:
        print(f"\nTotal tokens: {chunk.usage.total_tokens}")
    else:
        print(chunk.choices[0].delta.content or "", end="", flush=True)
```

## Next Steps

| Task                      | Guide                                                                 |
| :------------------------ | :-------------------------------------------------------------------- |
| Invoke External Functions | [Function Calling](/en/gateway/chat-completions/tool-calling)         |
| Validated JSON Outputs    | [Structured Outputs](/en/gateway/chat-completions/structured-outputs) |
| Extract Model Reasoning   | [Model Reasoning](/en/gateway/chat-completions/reasoning)             |


## Related topics

- [Chat Completions](/en/gateway/chat-completions.md)
- [Anthropic Streaming](/en/gateway/anthropic-messages/streaming.md)
- [Model Reasoning](/en/gateway/chat-completions/reasoning.md)
- [Rate Limits & Throughput](/en/guides/rate-limits.md)