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

# Responses API

> Referensi endpoint POST /v1/responses (OpenAI Responses API-compatible) - pakai Codex CLI, OpenAI Agents SDK, atau client.responses.create() dengan API key Hostlinku

Hostlinku menyediakan endpoint yang kompatibel dengan
[OpenAI Responses API](https://platform.openai.com/docs/api-reference/responses) - format yang
dipakai **Codex CLI**, **OpenAI Agents SDK**, dan metode `client.responses.create()` di SDK OpenAI.
Cukup ganti base URL; API key, saldo, Service Mode, rate limit, dan Request Logs sama dengan
`/v1/chat/completions` dan `/v1/messages`.

```http theme={null}
POST https://api.hostlinku.com/v1/responses
Authorization: Bearer sk-isi-api-key-kamu
Content-Type: application/json
```

Setiap respons menyertakan header `x-request-id` (ID request Hostlinku, format `resp_…`) dan, pada
respons sukses non-streaming, `x-model`.

Semua model, satu format

`model` memakai slug publik Hostlinku yang sama dengan [katalog model](https://docs.hostlinku.com/models.md) - mis.
`gpt-5-5`, `claude-sonnet-4-6`, `deepseek-v4-flash`. Model yang tidak menyediakan Responses API
secara native (termasuk Claude) tetap bisa dipanggil - gateway yang menerjemahkan.

## Konfigurasi cepat

Panduan lengkap per tool ada di [Codex CLI](https://docs.hostlinku.com/integrations/codex-cli.md) dan
[OpenAI Agents SDK](https://docs.hostlinku.com/integrations/agents-sdk.md).

**Codex CLI** (`~/.codex/config.toml`)

```toml theme={null}
model = "gpt-5-5"
model_provider = "hostlinku"

[model_providers.hostlinku]
name = "Hostlinku"
base_url = "https://api.hostlinku.com/v1"
env_key = "HOSTLINKU_API_KEY"
wire_api = "responses"
```

```bash theme={null}
export HOSTLINKU_API_KEY=sk-isi-api-key-kamu
codex
```

**Python (SDK resmi OpenAI)**

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

client = OpenAI(api_key="sk-isi-api-key-kamu", base_url="https://api.hostlinku.com/v1")

response = client.responses.create(
    model="gpt-5-5",
    instructions="Jawab singkat dalam bahasa Indonesia.",
    input="Apa itu API gateway?",
    max_output_tokens=256,
)
print(response.output_text)
print(response.usage.input_tokens, response.usage.output_tokens)
```

**Node.js**

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

const client = new OpenAI({ apiKey: "sk-isi-api-key-kamu", baseURL: "https://api.hostlinku.com/v1" });
const response = await client.responses.create({
  model: "claude-sonnet-4-6",
  input: "Halo!",
  max_output_tokens: 256,
});
console.log(response.output_text);
```

**OpenAI Agents SDK (Python)**

```python theme={null}
from agents import Agent, Runner, set_default_openai_client, set_tracing_disabled
from openai import AsyncOpenAI

set_default_openai_client(AsyncOpenAI(api_key="sk-isi-api-key-kamu", base_url="https://api.hostlinku.com/v1"))
set_tracing_disabled(True)  # tracing OpenAI tidak tersedia lewat gateway

agent = Agent(name="Asisten", instructions="Jawab singkat.", model="gpt-5-5")
result = Runner.run_sync(agent, "Sebutkan 3 bahasa pemrograman.")
print(result.final_output)
```

## Parameter body

| Parameter | Tipe | Keterangan |
| - | - | - |
| `model` | `string` | **Wajib.** Slug model Hostlinku. |
| `input` | `string` \| `array` | **Wajib.** Teks tunggal, atau array item: `message` (`role` + `content` berisi `input_text`/`input_image`/…), `function_call`, `function_call_output`, `reasoning`. |
| `instructions` | `string` | Prompt sistem (developer message). |
| `max_output_tokens` | `integer` | Batas token output. Ditolak `400` bila melebihi batas output model. |
| `stream` | `boolean` | `true` untuk SSE (lihat di bawah). |
| `tools`, `tool_choice`, `parallel_tool_calls` | - | Function calling format Responses (`{ "type": "function", "name", "parameters" }`). |
| `reasoning` | `object` | Mis. `{ "effort": "low" \| "medium" \| "high", "summary": "auto" }`. |
| `include` | `string[]` | Mis. `["reasoning.encrypted_content"]` - item reasoning terenkripsi dikembalikan apa adanya agar bisa dikirim ulang pada giliran berikutnya (dipakai Codex CLI). |
| `text`, `truncation`, `temperature`, `top_p`, `metadata`, `user`, `prompt_cache_key`, `safety_identifier` | - | Sesuai spesifikasi OpenAI; diteruskan ke model. |
| `store` | `boolean` | Selalu diperlakukan sebagai `false` (lihat *Stateless*). |
| `previous_response_id`, `conversation`, `background` | - | **Tidak didukung** → `400 invalid_request_error`. |

Field lain yang tidak dikenal diteruskan ke gateway (parameter yang tidak didukung model akan
diabaikan). Body tidak valid ditolak `400 invalid_request_error` dengan rincian per field di
`error.details.issues`.

Stateless - kirim riwayat penuh di `input`

Hostlinku **tidak menyimpan** response (`store: false`). Untuk percakapan multi-giliran, kirim
seluruh riwayat (termasuk item `function_call`, `function_call_output`, dan `reasoning` dari
respons sebelumnya) di array `input` - persis cara kerja Codex CLI. `previous_response_id`
ditolak dengan pesan yang jelas; `GET`/`DELETE /v1/responses/{id}` menjawab `404`.

## Respons non-streaming

```json theme={null}
{
  "id": "resp_…",
  "object": "response",
  "created_at": 1791850828,
  "completed_at": 1791850829,
  "status": "completed",
  "error": null,
  "incomplete_details": null,
  "model": "gpt-5-5",
  "output": [
    { "id": "rs_…", "type": "reasoning", "summary": [], "encrypted_content": "…" },
    {
      "id": "msg_…",
      "type": "message",
      "status": "completed",
      "role": "assistant",
      "content": [{ "type": "output_text", "text": "API gateway adalah …", "annotations": [] }]
    }
  ],
  "usage": {
    "input_tokens": 20,
    "output_tokens": 19,
    "total_tokens": 39,
    "input_tokens_details": { "cached_tokens": 0 },
    "output_tokens_details": { "reasoning_tokens": 12 }
  },
  "previous_response_id": null,
  "store": false,
  "hostlinku": { "cost": 0.00031, "currency": "🪙", "serviceMode": "performance", "latencyMs": 1420 }
}
```

`id` dan `model` selalu identitas publik Hostlinku. Blok `hostlinku` adalah ekstensi (SDK OpenAI
mengabaikannya) berisi biaya kredit, Service Mode API key, dan latensi. `status: "incomplete"` dengan
`incomplete_details.reason = "max_output_tokens"` berarti output terpotong oleh `max_output_tokens`

* tetap ditagih sesuai token nyata.

Model reasoning + `max_output_tokens` kecil

Pada model reasoning (mis. DeepSeek V4, GPT-6, Claude dengan thinking), token berpikir ikut
dihitung ke `max_output_tokens`. Nilai kecil (mis. `16`) biasanya habis untuk item `reasoning`
saja → `status: "incomplete"` dan `output` hanya berisi item `reasoning` tanpa `message`.
Gunakan `max_output_tokens` ≥ 256 atau turunkan `reasoning.effort` ke `"low"`.

Konvensi usage Responses API

Sama seperti Chat Completions, `input_tokens` **sudah memuat** token cache; rinciannya ada di
`input_tokens_details.cached_tokens`. `output_tokens` memuat `reasoning_tokens`. Tagihan Hostlinku
dihitung dari komponen tersebut dengan harga masing-masing (lihat [Billing](https://docs.hostlinku.com/billing.md)).

## Streaming (`stream: true`)

Gateway mengalirkan event SSE dengan framing Responses API (`event:` + `data:`), termasuk
`sequence_number` dari upstream:

```text theme={null}
event: response.created
data: {"type":"response.created","sequence_number":0,"response":{"id":"resp_…","object":"response","status":"in_progress","model":"gpt-5-5","output":[],…}}

event: response.in_progress
data: {"type":"response.in_progress","sequence_number":1,"response":{…}}

event: response.output_item.added
data: {"type":"response.output_item.added","sequence_number":2,"output_index":0,"item":{"id":"msg_…","type":"message","status":"in_progress","role":"assistant","content":[]}}

event: response.content_part.added
data: {"type":"response.content_part.added","sequence_number":3,"item_id":"msg_…","output_index":0,"content_index":0,"part":{"type":"output_text","text":"","annotations":[]}}

event: response.output_text.delta
data: {"type":"response.output_text.delta","sequence_number":4,"item_id":"msg_…","output_index":0,"content_index":0,"delta":"API gateway"}

event: response.output_text.done
data: {"type":"response.output_text.done","sequence_number":5,"item_id":"msg_…","output_index":0,"content_index":0,"text":"API gateway adalah …"}

event: response.content_part.done
data: {"type":"response.content_part.done","sequence_number":6,…}

event: response.output_item.done
data: {"type":"response.output_item.done","sequence_number":7,"output_index":0,"item":{"id":"msg_…","type":"message","status":"completed",…}}

event: response.completed
data: {"type":"response.completed","sequence_number":8,"response":{"id":"resp_…","status":"completed","model":"gpt-5-5","output":[…],"usage":{"input_tokens":20,"output_tokens":19,"total_tokens":39,…}}}
```

Catatan:

* Usage final ada di objek `response` pada `response.completed` (atau `response.incomplete`); biaya
  dihitung dari usage nyata setelah event tersebut.
* Tool call dialirkan lewat `response.output_item.added` (`type: "function_call"`),
  `response.function_call_arguments.delta`, dan `response.function_call_arguments.done`.
* Bila stream gagal di tengah, gateway mengirim `event: error` dengan
  `{"type":"error","code":…,"message":…}` dan **tidak** mengirim `response.completed` palsu.
* Error sebelum byte pertama (saldo habis, model tidak ada, rate limit) dikembalikan sebagai
  respons JSON biasa dengan status HTTP yang benar - SDK OpenAI memetakannya ke exception
  (`AuthenticationError`, `NotFoundError`, `RateLimitError`, …).

## Tool calling (function calling)

```python theme={null}
tools = [{
    "type": "function",
    "name": "get_time",
    "description": "Waktu saat ini di sebuah kota",
    "parameters": {"type": "object", "properties": {"city": {"type": "string"}}, "required": ["city"]},
}]

r1 = client.responses.create(model="claude-sonnet-4-6", input="Jam berapa sekarang di Jakarta?", tools=tools, store=False)
call = next(o for o in r1.output if o.type == "function_call")

# Giliran berikutnya: kirim ulang riwayat + hasil tool (stateless)
r2 = client.responses.create(
    model="claude-sonnet-4-6",
    tools=tools,
    store=False,
    input=[
        {"type": "message", "role": "user", "content": [{"type": "input_text", "text": "Jam berapa sekarang di Jakarta?"}]},
        *[o.model_dump(exclude_none=True) for o in r1.output],
        {"type": "function_call_output", "call_id": call.call_id, "output": '{"time": "10:00 WIB"}'},
    ],
)
print(r2.output_text)
```

## Format error

```json theme={null}
{ "error": { "message": "API key tidak valid atau sudah dicabut.", "type": "authentication_error", "param": null, "code": "invalid_api_key" } }
```

| HTTP | `error.type` | `error.code` | Penyebab umum |
| - | - | - | - |
| 400 | `invalid_request_error` | `invalid_request` | Body tidak valid, `max_output_tokens` melebihi batas model, `previous_response_id` dikirim, model tidak tersedia di Service Mode key. |
| 401 | `authentication_error` | `invalid_api_key` | API key salah/dicabut/kedaluwarsa, atau request tanpa API key. |
| 402 | `insufficient_quota` | `insufficient_quota` | Saldo tidak cukup atau limit kredit API key tercapai. |
| 403 | `permission_error` | - | Akun ditangguhkan / akses ditolak. |
| 404 | `not_found_error` | `model_not_found` / `not_found` | Slug model tidak ada; atau `GET /v1/responses/{id}` (stateless). |
| 429 | `rate_limit_error` | `rate_limit_exceeded` | Batas permintaan tercapai (lihat header `Retry-After`). |
| 503 | `server_error` | `model_unavailable` / `server_error` | Model/upstream sedang tidak tersedia - aman untuk di-retry. |

Pesan error tidak pernah memuat identitas provider upstream. Daftar lengkap kode untuk format
Chat Completions ada di [Error](https://docs.hostlinku.com/errors.md).


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