Utiliser l'API Messages

ConstruireConstruire avec Claude

Utilisation de l'API Messages

Modèles pratiques et exemples pour utiliser efficacement l'API Messages

Anthropic offers two ways to build with Claude, each suited to different use cases:

	Messages API	Claude Managed Agents
What it is	Direct model prompting access	Pre-built, configurable agent harness that runs in managed infrastructure
Best for	Custom agent loops and fine-grained control	Long-running tasks and asynchronous work
Learn more	Messages API docs	Claude Managed Agents docs

Ce guide couvre les modèles courants pour travailler avec l'API Messages, y compris les requêtes de base, les conversations multi-tours, les techniques de préremplissage et les capacités de vision. Pour les spécifications complètes de l'API, consultez la référence de l'API Messages.

This feature is eligible for Zero Data Retention (ZDR). When your organization has a ZDR arrangement, data sent through this feature is not stored after the API response is returned.

Requête et réponse de base

message = anthropic.Anthropic().messages.create(
    model="claude-opus-4-7",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello, Claude"}],
)
print(message)

Output

{
  "id": "msg_01XFDUDYJgAACzvnptvVoYEL",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "type": "text",
      "text": "Hello!"
    }
  ],
  "model": "claude-opus-4-7",
  "stop_reason": "end_turn",
  "stop_sequence": null,
  "usage": {
    "input_tokens": 12,
    "output_tokens": 6
  }
}

Tours de conversation multiples

L'API Messages est sans état, ce qui signifie que vous envoyez toujours l'historique conversationnel complet à l'API. Vous pouvez utiliser ce modèle pour construire une conversation au fil du temps. Les tours conversationnels antérieurs ne doivent pas nécessairement provenir réellement de Claude. Vous pouvez utiliser des messages assistant synthétiques.

message = anthropic.Anthropic().messages.create(
    model="claude-opus-4-7",
    max_tokens=1024,
    messages=[
        {"role": "user", "content": "Hello, Claude"},
        {"role": "assistant", "content": "Hello!"},
        {"role": "user", "content": "Can you describe LLMs to me?"},
    ],
)
print(message)

Output

{
  "id": "msg_018gCsTGsXkYJVqYPxTgDHBU",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "type": "text",
      "text": "Sure, I'd be happy to provide..."
    }
  ],
  "stop_reason": "end_turn",
  "stop_sequence": null,
  "usage": {
    "input_tokens": 30,
    "output_tokens": 309
  }
}

Mettre des paroles dans la bouche de Claude

Vous pouvez préremplir une partie de la réponse de Claude à la dernière position de la liste des messages d'entrée. Cela peut être utilisé pour façonner la réponse de Claude. L'exemple ci-dessous utilise "max_tokens": 1 pour obtenir une réponse à choix multiples unique de Claude.

message = anthropic.Anthropic().messages.create(
    model="claude-sonnet-4-5",
    max_tokens=1,
    messages=[
        {
            "role": "user",
            "content": "What is latin for Ant? (A) Apoidea, (B) Rhopalocera, (C) Formicidae",
        },
        {"role": "assistant", "content": "The answer is ("},
    ],
)
print(message)

Output

{
  "id": "msg_01Q8Faay6S7QPTvEUUQARt7h",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "type": "text",
      "text": "C"
    }
  ],
  "model": "claude-sonnet-4-5",
  "stop_reason": "max_tokens",
  "stop_sequence": null,
  "usage": {
    "input_tokens": 42,
    "output_tokens": 1
  }
}

Le préremplissage n'est pas pris en charge sur Claude Mythos Preview, Claude Opus 4.7, Claude Opus 4.6 et Claude Sonnet 4.6. Les requêtes utilisant le préremplissage avec ces modèles retournent une erreur 400. Utilisez plutôt les sorties structurées ou les instructions du message système. Consultez le guide de migration pour les modèles de migration.

Vision

Claude peut lire à la fois du texte et des images dans les demandes. Les images peuvent être fournies en utilisant les types de source base64, url ou file. Le type de source file fait référence à une image téléchargée via l'API Files. Les types de médias pris en charge sont image/jpeg, image/png, image/gif et image/webp. Consultez le guide de vision pour plus de détails.

import base64
import httpx

# Option 1: Base64-encoded image
image_url = "https://upload.wikimedia.org/wikipedia/commons/a/a7/Camponotus_flavomarginatus_ant.jpg"
image_media_type = "image/jpeg"
image_data = base64.standard_b64encode(httpx.get(image_url).content).decode("utf-8")

message = anthropic.Anthropic().messages.create(
    model="claude-opus-4-7",
    max_tokens=1024,
    messages=[
        {
            "role": "user",
            "content": [
                {
                    "type": "image",
                    "source": {
                        "type": "base64",
                        "media_type": image_media_type,
                        "data": image_data,
                    },
                },
                {"type": "text", "text": "What is in the above image?"},
            ],
        }
    ],
)
print(message)

# Option 2: URL-referenced image
message_from_url = anthropic.Anthropic().messages.create(
    model="claude-opus-4-7",
    max_tokens=1024,
    messages=[
        {
            "role": "user",
            "content": [
                {
                    "type": "image",
                    "source": {
                        "type": "url",
                        "url": "https://upload.wikimedia.org/wikipedia/commons/a/a7/Camponotus_flavomarginatus_ant.jpg",
                    },
                },
                {"type": "text", "text": "What is in the above image?"},
            ],
        }
    ],
)
print(message_from_url)

Output

{
  "id": "msg_01EcyWo6m4hyW8KHs2y2pei5",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "type": "text",
      "text": "This image shows an ant, specifically a close-up view of an ant. The ant is shown in detail, with its distinct head, antennae, and legs clearly visible. The image is focused on capturing the intricate details and features of the ant, likely taken with a macro lens to get an extreme close-up perspective."
    }
  ],
  "model": "claude-opus-4-7",
  "stop_reason": "end_turn",
  "stop_sequence": null,
  "usage": {
    "input_tokens": 1551,
    "output_tokens": 71
  }
}

Utilisation d'outils et utilisation d'ordinateur

Consultez le guide d'utilisation d'outils pour des exemples de la façon d'utiliser des outils avec l'API Messages. Consultez le guide d'utilisation d'ordinateur pour des exemples de la façon de contrôler les environnements de bureau informatique avec l'API Messages. Pour une sortie JSON garantie, consultez Structured Outputs.

Was this page helpful?

ConstruireConstruire avec Claude

Utilisation de l'API Messages

Modèles pratiques et exemples pour utiliser efficacement l'API Messages

Anthropic offers two ways to build with Claude, each suited to different use cases:

	Messages API	Claude Managed Agents
What it is	Direct model prompting access	Pre-built, configurable agent harness that runs in managed infrastructure
Best for	Custom agent loops and fine-grained control	Long-running tasks and asynchronous work
Learn more	Messages API docs	Claude Managed Agents docs

This feature is eligible for Zero Data Retention (ZDR). When your organization has a ZDR arrangement, data sent through this feature is not stored after the API response is returned.

Requête et réponse de base

message = anthropic.Anthropic().messages.create(
    model="claude-opus-4-7",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello, Claude"}],
)
print(message)

Output

{
  "id": "msg_01XFDUDYJgAACzvnptvVoYEL",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "type": "text",
      "text": "Hello!"
    }
  ],
  "model": "claude-opus-4-7",
  "stop_reason": "end_turn",
  "stop_sequence": null,
  "usage": {
    "input_tokens": 12,
    "output_tokens": 6
  }
}

Tours de conversation multiples

message = anthropic.Anthropic().messages.create(
    model="claude-opus-4-7",
    max_tokens=1024,
    messages=[
        {"role": "user", "content": "Hello, Claude"},
        {"role": "assistant", "content": "Hello!"},
        {"role": "user", "content": "Can you describe LLMs to me?"},
    ],
)
print(message)

Output

{
  "id": "msg_018gCsTGsXkYJVqYPxTgDHBU",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "type": "text",
      "text": "Sure, I'd be happy to provide..."
    }
  ],
  "stop_reason": "end_turn",
  "stop_sequence": null,
  "usage": {
    "input_tokens": 30,
    "output_tokens": 309
  }
}

Mettre des paroles dans la bouche de Claude

message = anthropic.Anthropic().messages.create(
    model="claude-sonnet-4-5",
    max_tokens=1,
    messages=[
        {
            "role": "user",
            "content": "What is latin for Ant? (A) Apoidea, (B) Rhopalocera, (C) Formicidae",
        },
        {"role": "assistant", "content": "The answer is ("},
    ],
)
print(message)

Output

{
  "id": "msg_01Q8Faay6S7QPTvEUUQARt7h",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "type": "text",
      "text": "C"
    }
  ],
  "model": "claude-sonnet-4-5",
  "stop_reason": "max_tokens",
  "stop_sequence": null,
  "usage": {
    "input_tokens": 42,
    "output_tokens": 1
  }
}

Vision

import base64
import httpx

# Option 1: Base64-encoded image
image_url = "https://upload.wikimedia.org/wikipedia/commons/a/a7/Camponotus_flavomarginatus_ant.jpg"
image_media_type = "image/jpeg"
image_data = base64.standard_b64encode(httpx.get(image_url).content).decode("utf-8")

message = anthropic.Anthropic().messages.create(
    model="claude-opus-4-7",
    max_tokens=1024,
    messages=[
        {
            "role": "user",
            "content": [
                {
                    "type": "image",
                    "source": {
                        "type": "base64",
                        "media_type": image_media_type,
                        "data": image_data,
                    },
                },
                {"type": "text", "text": "What is in the above image?"},
            ],
        }
    ],
)
print(message)

# Option 2: URL-referenced image
message_from_url = anthropic.Anthropic().messages.create(
    model="claude-opus-4-7",
    max_tokens=1024,
    messages=[
        {
            "role": "user",
            "content": [
                {
                    "type": "image",
                    "source": {
                        "type": "url",
                        "url": "https://upload.wikimedia.org/wikipedia/commons/a/a7/Camponotus_flavomarginatus_ant.jpg",
                    },
                },
                {"type": "text", "text": "What is in the above image?"},
            ],
        }
    ],
)
print(message_from_url)

Output

{
  "id": "msg_01EcyWo6m4hyW8KHs2y2pei5",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "type": "text",
      "text": "This image shows an ant, specifically a close-up view of an ant. The ant is shown in detail, with its distinct head, antennae, and legs clearly visible. The image is focused on capturing the intricate details and features of the ant, likely taken with a macro lens to get an extreme close-up perspective."
    }
  ],
  "model": "claude-opus-4-7",
  "stop_reason": "end_turn",
  "stop_sequence": null,
  "usage": {
    "input_tokens": 1551,
    "output_tokens": 71
  }
}

Utilisation d'outils et utilisation d'ordinateur

Was this page helpful?