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

# Stream tokens from abliteration.ai

> Stream abliteration.ai responses as server-sent events to reduce time-to-first-token.

**To stream tokens from abliteration.ai**, set `stream: true` on any chat completion request. The response is a sequence of [server-sent events](https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events) that render tokens as the model produces them — reducing time-to-first-token.

<Tabs>
  <Tab title="Python">
    ```python theme={"system"}
    from openai import OpenAI

    client = OpenAI(base_url="https://api.abliteration.ai/v1", api_key=os.environ["ABLIT_KEY"])

    stream = client.chat.completions.create(
        model="abliterated-model",
        messages=[{"role": "user", "content": "Write a haiku about streaming"}],
        stream=True,
    )

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

  <Tab title="Node">
    ```javascript theme={"system"}
    import OpenAI from "openai";

    const client = new OpenAI({
      baseURL: "https://api.abliteration.ai/v1",
      apiKey: process.env.ABLIT_KEY,
    });

    const stream = await client.chat.completions.create({
      model: "abliterated-model",
      messages: [{ role: "user", content: "Write a haiku about streaming" }],
      stream: true,
    });

    for await (const chunk of stream) {
      process.stdout.write(chunk.choices[0].delta.content ?? "");
    }
    ```
  </Tab>

  <Tab title="curl">
    ```sh theme={"system"}
    curl https://api.abliteration.ai/v1/chat/completions \
      -H "Authorization: Bearer $ABLIT_KEY" \
      -H "Content-Type: application/json" \
      -d '{
        "model": "abliterated-model",
        "messages": [{"role": "user", "content": "Write a haiku about streaming"}],
        "stream": true
      }'
    ```
  </Tab>
</Tabs>

<Note>
  Streamed chunks arrive as `data: {...}\n\n` SSE frames terminated by `data: [DONE]`. Most SDKs parse this for you.
</Note>

## Tool calls in streams

When the model calls a tool, `tool_calls` arrives across multiple chunks. Accumulate `function.arguments` string fragments until the chunk with `finish_reason: "tool_calls"`.

See [tool calling](/capabilities/tool-calling) for a complete example.