Anthropic Messages

Gunakan Anthropic Messages API melalui CometAPI untuk mengakses model Claude dengan extended thinking, prompt caching, tool use, web search/fetch, Streaming, dan effort control.

POST /v1/messages

Python

Basic

import anthropic

client = anthropic.Anthropic(
    base_url="https://api.cometapi.com",
    api_key="<COMETAPI_KEY>",
)

message = client.messages.create(
    model="claude-sonnet-4-6",
    max_tokens=1024,
    system="You are a helpful assistant.",
    messages=[
        {"role": "user", "content": "Hello, world"}
    ],
)

print(message.content[0].text)

200

{
  "id": "<string>",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "type": "text",
      "text": "<string>",
      "thinking": "<string>",
      "signature": "<string>",
      "id": "<string>",
      "name": "<string>",
      "input": {}
    }
  ],
  "model": "<string>",
  "stop_reason": "end_turn",
  "stop_sequence": "<string>",
  "usage": {
    "input_tokens": 123,
    "output_tokens": 123,
    "cache_creation_input_tokens": 123,
    "cache_read_input_tokens": 123,
    "cache_creation": {
      "ephemeral_5m_input_tokens": 123,
      "ephemeral_1h_input_tokens": 123
    }
  }
}

​

Gambaran Umum

CometAPI mendukung Anthropic Messages API secara native, memberi Anda akses langsung ke model Claude dengan semua fitur khusus Anthropic. Gunakan endpoint ini untuk kapabilitas eksklusif Claude seperti extended thinking, prompt caching, dan effort control.

​

Mulai Cepat

Gunakan SDK Anthropic resmi — cukup atur base URL ke CometAPI:

Python

import anthropic

client = anthropic.Anthropic(
    base_url="https://api.cometapi.com",
    api_key="<COMETAPI_KEY>",
)

message = client.messages.create(
    model="claude-sonnet-4-6",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello!"}],
)
print(message.content[0].text)

Header x-api-key dan Authorization: Bearer sama-sama didukung untuk autentikasi. SDK Anthropic resmi menggunakan x-api-key secara default.

​

Extended Thinking

Aktifkan penalaran langkah demi langkah Claude dengan parameter thinking. Respons mencakup blok konten thinking yang menampilkan penalaran internal Claude sebelum jawaban akhir.

message = client.messages.create(
    model="claude-sonnet-4-6",
    max_tokens=16000,
    thinking={
        "type": "enabled",
        "budget_tokens": 10000,
    },
    messages=[
        {"role": "user", "content": "Prove that there are infinitely many primes."}
    ],
)

for block in message.content:
    if block.type == "thinking":
        print(f"Thinking: {block.thinking[:200]}...")
    elif block.type == "text":
        print(f"Answer: {block.text}")

Thinking memerlukan budget_tokens minimum sebesar 1,024. Thinking tokens dihitung ke dalam batas max_tokens Anda — atur max_tokens cukup tinggi untuk menampung thinking dan respons.

---

​

Prompt Caching

Cache prompt system yang besar atau prefix percakapan untuk mengurangi latensi dan biaya pada request berikutnya. Tambahkan cache_control ke blok konten yang ingin di-cache:

message = client.messages.create(
    model="claude-sonnet-4-6",
    max_tokens=1024,
    system=[
        {
            "type": "text",
            "text": "You are an expert code reviewer. [Long detailed instructions...]",
            "cache_control": {"type": "ephemeral"},
        }
    ],
    messages=[{"role": "user", "content": "Review this code..."}],
)

Penggunaan cache dilaporkan dalam field usage pada respons:

• cache_creation_input_tokens — token yang ditulis ke cache (ditagih dengan tarif lebih tinggi)
• cache_read_input_tokens — token yang dibaca dari cache (ditagih dengan tarif lebih rendah)

Prompt caching memerlukan minimum 1,024 tokens dalam blok konten yang di-cache. Konten yang lebih pendek dari ini tidak akan di-cache.

---

​

Streaming

Stream respons menggunakan Server-Sent Events (SSE) dengan menetapkan stream: true. Event datang dalam urutan berikut:

1. message_start — berisi metadata message dan usage awal
2. content_block_start — menandai awal setiap blok content
3. content_block_delta — potongan teks bertahap (text_delta)
4. content_block_stop — menandai akhir setiap blok content
5. message_delta — stop_reason final dan usage lengkap
6. message_stop — menandakan akhir stream

with client.messages.stream(
    model="claude-sonnet-4-6",
    max_tokens=256,
    messages=[{"role": "user", "content": "Hello"}],
) as stream:
    for text in stream.text_stream:
        print(text, end="")

---

​

Kontrol Effort

Atur seberapa besar effort yang Claude gunakan untuk menghasilkan respons dengan output_config.effort:

message = client.messages.create(
    model="claude-opus-4-6",
    max_tokens=4096,
    messages=[
        {"role": "user", "content": "Summarize this briefly."}
    ],
    output_config={"effort": "low"},  # "low", "medium", or "high"
)

---

​

Server Tools

Claude mendukung server-side tools yang berjalan di infrastruktur Anthropic:

Web Fetch

Ambil dan analisis content dari URL:

message = client.messages.create(
    model="claude-sonnet-4-6",
    max_tokens=1024,
    messages=[
        {"role": "user", "content": "Analyze the content at https://arxiv.org/abs/1512.03385"}
    ],
    tools=[
        {"type": "web_fetch_20250910", "name": "web_fetch", "max_uses": 5}
    ],
)

Web Search

Telusuri web untuk informasi real-time:

message = client.messages.create(
    model="claude-sonnet-4-6",
    max_tokens=1024,
    messages=[
        {"role": "user", "content": "What are the latest developments in AI?"}
    ],
    tools=[
        {"type": "web_search_20250305", "name": "web_search", "max_uses": 5}
    ],
)

---

​

Contoh Respons

Respons umum dari endpoint Anthropic CometAPI:

{
  "id": "msg_bdrk_01UjHdmSztrL7QYYm7CKBDFB",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "type": "text",
      "text": "Hello!"
    }
  ],
  "model": "claude-sonnet-4-6",
  "stop_reason": "end_turn",
  "stop_sequence": null,
  "usage": {
    "input_tokens": 19,
    "cache_creation_input_tokens": 0,
    "cache_read_input_tokens": 0,
    "cache_creation": {
      "ephemeral_5m_input_tokens": 0,
      "ephemeral_1h_input_tokens": 0
    },
    "output_tokens": 4
  }
}

---

​

Perbedaan Utama dari Endpoint yang Kompatibel dengan OpenAI

Fitur	Anthropic Messages (/v1/messages)	OpenAI-Compatible (/v1/chat/completions)
Extended thinking	parameter thinking dengan budget_tokens	Tidak tersedia
Prompt caching	cache_control pada blok content	Tidak tersedia
Kontrol effort	output_config.effort	Tidak tersedia
Web fetch/search	Server tools (web_fetch, web_search)	Tidak tersedia
Header auth	x-api-key atau Bearer	Hanya Bearer
Format respons	format Anthropic (content blocks)	format OpenAI (choices, message)
Model	Claude saja	Multi-provider (GPT, Claude, Gemini, dll.)

Otorisasi

x-api-key · string · header · wajib

Your CometAPI key passed via the x-api-key header. Authorization: Bearer <key> is also supported.

Header

anthropic-version · string · default: 2023-06-01

The Anthropic API version to use. Defaults to 2023-06-01.

Contoh:

"2023-06-01"

anthropic-beta · string

Comma-separated list of beta features to enable. Examples: max-tokens-3-5-sonnet-2024-07-15, pdfs-2024-09-25, output-128k-2025-02-19.

Body

application/json

model · string · wajib

The Claude model to use. See the Models page for current Claude model IDs.

Contoh:

"claude-sonnet-4-6"

messages · object[] · wajib

The conversation messages. Must alternate between user and assistant roles. Each message's content can be a string or an array of content blocks (text, image, document, tool_use, tool_result). There is a limit of 100,000 messages per request.

Show child attributes

max_tokens · integer · wajib

The maximum number of tokens to generate. The model may stop before reaching this limit. When using thinking, the thinking tokens count towards this limit.

Rentang yang diperlukan: x >= 1

Contoh:

1024

system · string object[]

System prompt providing context and instructions to Claude. Can be a plain string or an array of content blocks (useful for prompt caching).

temperature · number · default: 1

Controls randomness in the response. Range: 0.0–1.0. Use lower values for analytical tasks and higher values for creative tasks. Defaults to 1.0.

Rentang yang diperlukan: 0 <= x <= 1

top_p · number

Nucleus sampling threshold. Only tokens with cumulative probability up to this value are considered. Range: 0.0–1.0. Use either temperature or top_p, not both.

Rentang yang diperlukan: 0 <= x <= 1

top_k · integer

Only sample from the top K most probable tokens. Recommended for advanced use cases only.

Rentang yang diperlukan: x >= 0

stream · boolean · default: false

If true, stream the response incrementally using Server-Sent Events (SSE). Events include message_start, content_block_start, content_block_delta, content_block_stop, message_delta, and message_stop.

stop_sequences · string[]

Custom strings that cause the model to stop generating when encountered. The stop sequence is not included in the response.

thinking · object

Enable extended thinking — Claude's step-by-step reasoning process. When enabled, the response includes thinking content blocks before the answer. Requires a minimum budget_tokens of 1,024.

Show child attributes

tools · object[]

Tools the model may use. Supports client-defined functions, web search (web_search_20250305), web fetch (web_fetch_20250910), code execution (code_execution_20250522), and more.

Show child attributes

tool_choice · object

Controls how the model uses tools.

Show child attributes

metadata · object

Request metadata for tracking and analytics.

Show child attributes

output_config · object

Configuration for output behavior.

Show child attributes

service_tier · enum<string>

The service tier to use. auto tries priority capacity first, standard_only uses only standard capacity.

Opsi yang tersedia: auto, standard_only

Respons

200 - application/json

Successful response. When stream is true, the response is a stream of SSE events.

id · string

Unique identifier for this message (e.g., msg_01XFDUDYJgAACzvnptvVoYEL).

type · enum<string>

Always message.

Opsi yang tersedia: message

role · enum<string>

Always assistant.

Opsi yang tersedia: assistant

content · object[]

The response content blocks. May include text, thinking, tool_use, and other block types.

Show child attributes

model · string

The specific model version that generated this response (e.g., claude-sonnet-4-6).

stop_reason · enum<string>

Why the model stopped generating.

Opsi yang tersedia: end_turn, max_tokens, stop_sequence, tool_use, pause_turn

stop_sequence · string | null

The stop sequence that caused the model to stop, if applicable.

usage · object

Token usage statistics.

Show child attributes