Buat Message

RingkasanCopy Anchor Link

Endpoint ini menyediakan kompatibilitas native dengan Anthropic Messages API. Gunakan ini untuk model Claude dengan fitur seperti extended thinking.

Endpoint ini tetap mengikuti kontrak native Anthropic. messages harus berupa array pesan user / assistant, system berada pada field top-level system, dan max_tokens wajib dikirim. Jika payload menggunakan role OpenAI seperti system, developer, atau tool di dalam messages, kirim ke /v1/chat/completions.

Header PermintaanCopy Anchor Link

Body PermintaanCopy Anchor Link

ResponsCopy Anchor Link

curl -X POST "https://api.tokenlab.sh/v1/messages" \
  -H "x-api-key: sk-your-api-key" \
  -H "anthropic-version: 2023-06-01" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-5",
    "max_tokens": 1024,
    "system": "You are a helpful assistant.",
    "messages": [
      {"role": "user", "content": "Hello, Claude!"}
    ]
  }'

{
  "id": "msg_abc123",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "type": "text",
      "text": "Hello! How can I help you today?"
    }
  ],
  "model": "claude-sonnet-5",
  "stop_reason": "end_turn",
  "usage": {
    "input_tokens": 15,
    "output_tokens": 10
  }
}

Contoh Input VisualCopy Anchor Link

Untuk model Claude dengan dukungan visi, letakkan gambar di dalam messages[].content sebagai blok gambar terstruktur.

{
  "model": "claude-sonnet-5",
  "max_tokens": 1024,
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "Please describe this image."
        },
        {
          "type": "image",
          "source": {
            "type": "url",
            "url": "https://example.com/demo.jpg"
          }
        }
      ]
    }
  ]
}

{
  "model": "claude-sonnet-5",
  "max_tokens": 1024,
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "Please describe this image."
        },
        {
          "type": "image",
          "source": {
            "type": "base64",
            "media_type": "image/jpeg",
            "data": "/9j/4AAQSkZJRgABAQ..."
          }
        }
      ]
    }
  ]
}

Contoh Extended ThinkingCopy Anchor Link

Atur TOKENLAB_THINKING_MODEL ke model yang mendukung thinking manual dengan type: "enabled" dan budget_tokens. Sesuaikan kedua anggaran token dengan batas model tersebut. Model lain mungkin memerlukan konfigurasi thinking yang berbeda.

import os
from anthropic import Anthropic

client = Anthropic(
    api_key="sk-your-api-key",
    base_url="https://api.tokenlab.sh"
)

message = client.messages.create(
    model=os.environ["TOKENLAB_THINKING_MODEL"],
    max_tokens=16000,
    thinking={
        "type": "enabled",
        "budget_tokens": 10000
    },
    messages=[{"role": "user", "content": "Solve this math problem..."}]
)

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

Anthropic Batch PesanCopy Anchor Link

TokenLab kini juga menyediakan alur native Anthropic Message Batches di samping /v1/messages.

Rute yang tersedia:

• POST /v1/messages/batches
• GET /v1/messages/batches
• GET /v1/messages/batches/:message_batch_id
• GET /v1/messages/batches/:message_batch_id/results
• POST /v1/messages/batches/:message_batch_id/cancel
• DELETE /v1/messages/batches/:message_batch_id

Catatan operasional:

• Gunakan API key TokenLab yang sama bersama header native Anthropic.
• Jika item batch mereferensikan file_id, sertakan juga anthropic-beta: files-api-2025-04-14.
• Batch job mempertahankan bentuk request/response native Anthropic, sementara TokenLab melacak status billing untuk rekonsiliasi.