# Streaming

Set `"stream": true`. Respons dikirim sebagai Server-Sent Events (SSE), persis seperti OpenAI /
Anthropic, jadi semua SDK langsung bisa dipakai.

```bash
curl -N https://ai.brandkamu.id/v1/chat/completions \
  -H "Authorization: Bearer sk-..." \
  -H "Content-Type: application/json" \
  -d '{"model": "gpt-5-mini", "stream": true,
       "messages": [{"role": "user", "content": "Hitung 1 sampai 10"}]}'
```

```python
stream = client.chat.completions.create(
    model="gpt-5-mini",
    messages=[{"role": "user", "content": "Hitung 1 sampai 10"}],
    stream=True,
)
for chunk in stream:
    if chunk.choices and chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="", flush=True)
```

Catatan:

- SumoRouter otomatis meminta `usage` di akhir stream supaya biaya bisa dihitung; chunk terakhir berisi jumlah token.
- Jika koneksi terputus di tengah jalan, token yang sudah dihasilkan **tetap ditagih**.
- Di belakang reverse proxy sendiri, matikan buffering (SumoRouter sudah mengirim `X-Accel-Buffering: no`).
