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

# Antworten

> Rufe die Responses API mit dem TypeScript-SDK auf.

**Methode**: `client.generateResponse()` (übertrage verarbeitete Chunks mit `client.streamResponses()`).

### Beispiel

```typescript theme={null}
const resp = await client.generateResponse({
  model: "openai/gpt-4.1",
  input: [{ role: "user", content: [{ type: "input_text", text: "Summarise this text" }] }],
  temperature: 0.7,
  stream: false,
});
```

### Datenstrom

```typescript theme={null}
for await (const chunk of client.streamResponses({
  model: "openai/gpt-4.1",
  input: [{ role: "user", content: [{ type: "input_text", text: "Stream this" }] }],
  stream: true,
})) {
  if (chunk.text) process.stdout.write(chunk.text);
}
```

### Wichtige Parameter

* `model` (erforderlich): ID des Zielmodells.
* `input` (erforderlich): Geordnetes Array von Eingabeelementen (Nachrichten, Tool-Aufrufe usw.).
* `temperature` (0-2): Ein höherer Wert erhöht die Zufälligkeit.
* `top_p` (0-1) / `top_k` (>=1): Parameter für Nucleus-/Top-k-Sampling.
* `max_output_tokens` (int): Obergrenze für die pro Antwort generierten Tokens.
* Werkzeuge: `tools` (Definitionen), `tool_choice` (`auto`/`none`/spezifisches Werkzeug), `max_tool_calls` (int), `parallel_tool_calls` (bool).
* Logprobs: `logprobs` (bool), `top_logprobs` (0-20) um Logprobs pro Token zurückzugeben.
* Ausgabe: `response_format` (json/text), `service_tier`, `store` (bool), `stream` (bool).
* Metadaten: `metadata` (object) zur unveränderten Weitergabe, `reasoning` (object) als Hinweis zum Denkaufwand.
* Zusätzliche Gateway-Optionen: `usage` (bool, um Nutzungsdaten anzufordern), `meta` (bool, um den Meta-Block einzuschließen).

### Rückgabe

`ResponsesResponse`

```json theme={null}
{
  "id": "resp_123",
  "object": "response",
  "created_at": 1677652288,
  "status": "completed",
  "model": "gpt-4.1",
  "output": [
    {
      "type": "message",
      "id": "msg_123",
      "status": "completed",
      "role": "assistant",
      "content": [
        {
          "type": "output_text",
          "text": "Hello there, how may I assist you today?"
        }
      ]
    }
  ],
  "usage": {
    "input_tokens": 9,
    "output_tokens": 12,
    "total_tokens": 21
  }
}
```

Oder zeilengetrennte SSE-Frames bei `stream: true`:

```
data: {"id":"resp_123","object":"response","created_at":1677652288,"status":"in_progress","model":"gpt-4.1","output":[{"type":"message","id":"msg_123","status":"in_progress","role":"assistant","content":[]}]}

data: {"id":"resp_123","object":"response","created_at":1677652288,"status":"in_progress","model":"gpt-4.1","output":[{"type":"message","id":"msg_123","status":"in_progress","role":"assistant","content":[{"type":"output_text","text":"Hello"}]}]}

data: {"id":"resp_123","object":"response","created_at":1677652288,"status":"completed","model":"gpt-4.1","output":[{"type":"message","id":"msg_123","status":"completed","role":"assistant","content":[{"type":"output_text","text":"Hello there, how may I assist you today?"}]}],"usage":{"input_tokens":9,"output_tokens":12,"total_tokens":21}}

data: [DONE]
```


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.