Claude Messages | Dokumentasi API

# Claude Messages

Native Claude Messages-compatible API. This endpoint uses Anthropic Messages request and response shapes, including tool calls and Anthropic SSE events.

POST https://api.apihop.ai/anthropic/v1/messages

Native Claude Messages-compatible API. This endpoint uses Anthropic Messages request and response shapes, including tool calls and Anthropic SSE events.

Claude Sonnet 4.6 (`claude-sonnet-4-6`) is supported by the upstream supplier but is not currently published on API Hop. It is not an active integration default. Before calling any model, confirm access with your own `GET https://api.apihop.ai/v1/models` response.

```text
POST https://api.apihop.ai/anthropic/v1/messages
```

## Authentication

All requests require a API Hop API Key. Anthropic SDKs, Claude Code, and compatible clients usually use `x-api-key`; plain HTTP clients may also use Bearer Token authentication.

```http
x-api-key: YOUR_APIHOP_API_KEY
anthropic-version: 2023-06-01
```

```http
Authorization: Bearer YOUR_APIHOP_API_KEY
anthropic-version: 2023-06-01
```

## Body

### `model` (string, required, Default: claude-sonnet-5-5)

Supported Claude model IDs.

- `claude-fable-5`
- `claude-fable-5-1`
- `claude-opus-4-6`
- `claude-opus-4-7`
- `claude-opus-4-8`
- `claude-opus-5`
- `claude-opus-5-5`
- `claude-sonnet-5`
- `claude-sonnet-5-5`

Use `GET https://api.apihop.ai/v1/models` for the models currently available to your API Key.

### `max_tokens` (integer, required)

Maximum number of output tokens, as a positive safe integer. This budget includes both thinking and final text output tokens. Missing, zero, negative, fractional, string, or unsafe-integer values return HTTP `400` with `invalid_request_error` before generation or billing. The model may impose a lower maximum.

For models that produce thinking, a small budget such as `128` may yield only thinking blocks before ending with `stop_reason: "max_tokens"`, without text output. Set an explicit budget for the task and the selected model, allowing room for thinking and the final answer. The values in the examples are editable allowances, not upstream defaults or model limits; larger allowances may increase output usage and cost.

### `messages` (object[], required)

Conversation messages. Each message contains `role` and `content`.

`role` supports `user` and `assistant`. `content` may be a string or an array of content blocks.

### `system` (string | object[])

System prompt. Anthropic Messages uses a top-level `system` field instead of a `system` role in `messages`.

### `stream` (boolean, Default: false)

Whether to return Anthropic SSE events.

### `temperature` (number)

Sampling temperature. Accepted values and effects vary by the selected model.

### `top_p` (number)

Nucleus sampling parameter. Acceptance does not guarantee that it changes generation behavior.

### `top_k` (integer)

Top-k sampling parameter. Support depends on the selected model.

### `stop_sequences` (string[])

Stop sequences.

Requests with nonempty stop sequences avoid model services known not to support them; if no compatible service is available, the request returns `400` before generation or billing.

### `tools` (object[])

Anthropic tool definitions. Function tools usually include `name`, `description`, and `input_schema`.

### `tool_choice` (object)

Tool selection policy, such as `{ "type": "auto" }` or `{ "type": "tool", "name": "..." }`.

### `thinking` (object)

Native Anthropic thinking configuration, subject to the selected model and provider's support. Claude 4.5 uses `{ "type": "enabled", "budget_tokens": 1024 }`; Claude Sonnet/Opus 4.6 also support adaptive thinking. Do not apply a manual budget to every model version. In tool continuations, return the complete ordered assistant content, including `thinking`, `signature`, `redacted_thinking` and `tool_use`, unchanged. Keep the thinking configuration and required `anthropic-beta` headers consistent through the tool round trip. For SSE, concatenate `signature_delta.signature` fragments for the same content block; an empty thinking summary does not mean its signature can be discarded. A provider account access rejection is an upstream failure, not evidence of an invalid signature. Recognized Bedrock account access rejections return HTTP 502 with `api_error`; ordinary client validation errors retain HTTP 400. Acceptance still requires an enabled upstream account.

### `metadata` (object)

Request metadata.

### Prompt Caching

Add `cache_control: { "type": "ephemeral" }` to a `system` or message content block. Cache usage may include `cache_creation_input_tokens`, `cache_read_input_tokens`, and supported 5-minute or 1-hour creation buckets. A 1-hour TTL also requires the corresponding `anthropic-beta` header.

## Request Examples

### Non-Streaming

```json
{
  "model": "claude-sonnet-5-5",
  "max_tokens": 1024,
  "messages": [
    {
      "role": "user",
      "content": "Hello, introduce yourself in one sentence."
    }
  ],
  "stream": false
}
```

### Streaming

```json
{
  "model": "claude-sonnet-5-5",
  "max_tokens": 1024,
  "messages": [
    {
      "role": "user",
      "content": "Reply exactly OK."
    }
  ],
  "stream": true
}
```

## Response

Non-streaming responses keep the Anthropic Message shape.

### `id` (string)

Response ID, such as `msg_...`.

### `type` (string)

Object type, usually `message`.

### `role` (string)

Usually `assistant`.

### `model` (string)

Actual Claude model used.

### `content` (object[])

Native output content blocks, including `text`, `tool_use`, and `thinking` when used.

### `stop_reason` (string)

Stop reason, such as `end_turn`, `max_tokens`, `stop_sequence`, or `tool_use`.

### `usage` (object)

Token usage, including `input_tokens` and `output_tokens`. Cache token fields are included when present.

### Success Example

```json
{
  "id": "msg_01...",
  "type": "message",
  "role": "assistant",
  "model": "claude-sonnet-5-5",
  "content": [
    {
      "type": "text",
      "text": "Hello, I am Claude and can help with writing, coding, and analysis tasks."
    }
  ],
  "stop_reason": "end_turn",
  "stop_sequence": null,
  "usage": {
    "input_tokens": 18,
    "output_tokens": 24
  }
}
```

## Streaming Response

When `stream: true`, the endpoint returns Anthropic SSE events. Common events include:

- `message_start`
- `content_block_start`
- `content_block_delta`
- `content_block_stop`
- `ping`
- `message_delta`
- `message_stop`

## Error Response

See [Error Codes and Troubleshooting](https://apihop.ai/en/docs/api-reference/errors) for the full code reference, protocol differences, and diagnostic steps.

Error responses use the Anthropic-style shape:

```json
{
  "type": "error",
  "error": {
    "type": "authentication_error",
    "message": "Invalid credentials or API key has expired."
  }
}
```

API native yang kompatibel dengan Claude Messages. Endpoint ini menggunakan bentuk permintaan dan respons Anthropic Messages, termasuk panggilan alat dan peristiwa SSE Anthropic.

Claude Sonnet 4.6 (claude-sonnet-4-6) didukung penyedia hulu tetapi saat ini tidak dipublikasikan di API Hop. Model ini bukan pilihan integrasi bawaan yang aktif. Sebelum memanggil model apa pun, pastikan akses melalui respons GET https://api.apihop.ai/v1/models Anda sendiri.

text

POST https://api.apihop.ai/anthropic/v1/messages

Autentikasi

Semua permintaan memerlukan API Key API Hop. SDK Anthropic, Claude Code, dan klien kompatibel biasanya menggunakan x-api-key; klien HTTP biasa juga dapat menggunakan autentikasi Bearer Token.

http

x-api-key: YOUR_APIHOP_API_KEY
anthropic-version: 2023-06-01

http

Authorization: Bearer YOUR_APIHOP_API_KEY
anthropic-version: 2023-06-01

Isi

modelstringwajibBawaan: claude-sonnet-5-5

ID model Claude yang didukung.

claude-fable-5
claude-fable-5-1
claude-opus-4-6
claude-opus-4-7
claude-opus-4-8
claude-opus-5
claude-opus-5-5
claude-sonnet-5
claude-sonnet-5-5

Gunakan GET https://api.apihop.ai/v1/models untuk model yang saat ini tersedia bagi API Key Anda.

max_tokensintegerwajib

Jumlah maksimum token keluaran, berupa bilangan bulat aman positif. Anggaran ini mencakup token keluaran pemikiran dan teks akhir. Nilai yang tidak ada, nol, negatif, pecahan, string, atau bilangan bulat tidak aman mengembalikan HTTP 400 dengan invalid_request_error sebelum pembuatan atau penagihan. Model dapat menetapkan maksimum yang lebih rendah.

Untuk model yang menghasilkan pemikiran, anggaran kecil seperti 128 mungkin hanya menghasilkan blok pemikiran sebelum berakhir dengan stop_reason: "max_tokens", tanpa keluaran teks. Atur anggaran eksplisit untuk tugas dan model yang dipilih, dengan ruang untuk pemikiran serta jawaban akhir. Nilai dalam contoh adalah alokasi yang dapat disunting, bukan nilai bawaan upstream atau batas model; alokasi lebih besar dapat meningkatkan penggunaan keluaran dan biaya.

messagesobject[]wajib

Pesan percakapan. Setiap pesan berisi role dan content.

role mendukung user dan assistant. content dapat berupa string atau larik blok konten.

systemstring | object[]

Prompt sistem. Anthropic Messages menggunakan kolom system tingkat atas, bukan peran system dalam messages.

streambooleanBawaan: false

Apakah mengembalikan peristiwa SSE Anthropic.

temperaturenumber

Temperatur sampling. Nilai yang diterima dan efeknya berbeda menurut model yang dipilih.

top_pnumber

Parameter nucleus sampling. Penerimaan tidak menjamin perubahan perilaku pembuatan.

top_kinteger

Parameter sampling top-k. Dukungan bergantung pada model yang dipilih.

stop_sequencesstring[]

Urutan penghenti.

Permintaan dengan urutan penghenti yang tidak kosong menghindari layanan model yang diketahui tidak mendukungnya; jika tidak ada layanan kompatibel, permintaan mengembalikan 400 sebelum pembuatan atau penagihan.

toolsobject[]

Definisi alat Anthropic. Alat fungsi biasanya mencakup name, description, dan input_schema.

tool_choiceobject

Kebijakan pemilihan alat, seperti { "type": "auto" } atau { "type": "tool", "name": "..." }.

thinkingobject

Konfigurasi pemikiran native Anthropic, bergantung pada dukungan model dan penyedia yang dipilih. Claude 4.5 menggunakan { "type": "enabled", "budget_tokens": 1024 }; Claude Sonnet/Opus 4.6 juga mendukung pemikiran adaptif. Jangan menerapkan anggaran manual ke setiap versi model. Dalam kelanjutan alat, kembalikan konten assistant lengkap sesuai urutan, termasuk thinking, signature, redacted_thinking dan tool_use, tanpa perubahan. Pertahankan konfigurasi pemikiran dan header anthropic-beta yang diperlukan secara konsisten sepanjang siklus alat. Untuk SSE, gabungkan fragmen signature_delta.signature untuk blok konten yang sama; ringkasan pemikiran kosong tidak berarti tanda tangannya boleh dibuang. Penolakan akses akun penyedia adalah kegagalan upstream, bukan bukti tanda tangan tidak valid. Penolakan akses akun Bedrock yang dikenali mengembalikan HTTP 502 dengan api_error; kesalahan validasi klien biasa tetap menggunakan HTTP 400. Penerimaan tetap memerlukan akun upstream yang diaktifkan.

metadataobject

Metadata permintaan.

Cache Prompt

Tambahkan cache_control: { "type": "ephemeral" } ke blok konten system atau pesan. Penggunaan cache dapat mencakup cache_creation_input_tokens, cache_read_input_tokens, dan kategori pembuatan 5 menit atau 1 jam yang didukung. TTL 1 jam juga memerlukan header anthropic-beta yang sesuai.

Contoh Permintaan

Tanpa Streaming

json

{
  "model": "claude-sonnet-5-5",
  "max_tokens": 1024,
  "messages": [
    {
      "role": "user",
      "content": "Hello, introduce yourself in one sentence."
    }
  ],
  "stream": false
}

Respons streaming

json

{
  "model": "claude-sonnet-5-5",
  "max_tokens": 1024,
  "messages": [
    {
      "role": "user",
      "content": "Reply exactly OK."
    }
  ],
  "stream": true
}

Respons

Respons tanpa streaming mempertahankan bentuk Anthropic Message.

idstring

ID respons, seperti msg_....

typestring

Tipe objek, biasanya message.

rolestring

Biasanya assistant.

modelstring

Model Claude yang sebenarnya digunakan.

contentobject[]

Blok konten keluaran native, termasuk text, tool_use, dan thinking saat digunakan.

stop_reasonstring

Alasan penghentian, seperti end_turn, max_tokens, stop_sequence, atau tool_use.

usageobject

Penggunaan token, termasuk input_tokens dan output_tokens. Kolom token cache disertakan jika ada.

Contoh Berhasil

json

{
  "id": "msg_01...",
  "type": "message",
  "role": "assistant",
  "model": "claude-sonnet-5-5",
  "content": [
    {
      "type": "text",
      "text": "Hello, I am Claude and can help with writing, coding, and analysis tasks."
    }
  ],
  "stop_reason": "end_turn",
  "stop_sequence": null,
  "usage": {
    "input_tokens": 18,
    "output_tokens": 24
  }
}

Respons Streaming

Saat stream: true, endpoint mengembalikan peristiwa SSE Anthropic. Peristiwa umum meliputi:

message_start
content_block_start
content_block_delta
content_block_stop
ping
message_delta
message_stop

Respons Kesalahan

Lihat Kode Kesalahan dan Pemecahan Masalah untuk referensi kode lengkap, perbedaan protokol, dan langkah diagnosis.

Respons kesalahan menggunakan bentuk bergaya Anthropic:

json

{
  "type": "error",
  "error": {
    "type": "authentication_error",
    "message": "Invalid credentials or API key has expired."
  }
}