---
title: Contagem de tokens
url: https://platform.claude.com/docs/pt-BR/build-with-claude/token-counting
description: Conte os tokens em uma mensagem antes de enviá-la ao Claude. Use contagens de tokens para gerenciar limites de taxa e custos, tomar decisões de roteamento de modelos e ajustar prompts a um comprimento desejado.
featureMetadata:
  status: ga
  zdr:
    eligibility: eligible
    note: Excludes [Covered Models](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#model-specific-data-retention-requirements).
  supportedPlatforms:
    Claude API: ga
    Claude Platform on AWS: ga
    Amazon Bedrock: ga
    Google Cloud: ga
    Microsoft Foundry: ga
---

O "token counting" (contagem de tokens) permite determinar o número de tokens em uma mensagem antes de enviá-la ao Claude. Isso ajuda você a tomar decisões informadas sobre seus prompts e seu uso. Com a contagem de tokens, você pode:

* Gerenciar proativamente "rate limits" (limites de taxa) e custos
* Tomar decisões inteligentes de roteamento de modelos
* Otimizar prompts para um comprimento específico

***

## Como contar os tokens de uma mensagem

O endpoint de [contagem de tokens](https://platform.claude.com/docs/pt-BR/api/messages/count_tokens) aceita a mesma lista estruturada de entradas usada para criar uma mensagem, incluindo suporte a "system prompts" (prompts do sistema), [ferramentas](https://platform.claude.com/docs/pt-BR/agents-and-tools/tool-use/overview), [imagens](https://platform.claude.com/docs/pt-BR/build-with-claude/vision) e [PDFs](https://platform.claude.com/docs/pt-BR/build-with-claude/pdf-support). A resposta contém o número total de tokens de entrada.

Este endpoint retorna um `invalid_request_error` para algumas entradas que a Messages API aceita: [ferramentas de servidor](https://platform.claude.com/docs/pt-BR/agents-and-tools/tool-use/server-tools), como busca na web, busca de conteúdo da web, execução de código e busca de ferramentas (todas as ferramentas de servidor, exceto a [ferramenta advisor](https://platform.claude.com/docs/pt-BR/agents-and-tools/tool-use/advisor-tool)), o [conector MCP](https://platform.claude.com/docs/pt-BR/agents-and-tools/mcp-connector) e blocos `image` ou `document` com uma origem `url` ou `file`. Envie imagens e PDFs em base64 para contá-los. Para requisições que usam ferramentas de servidor ou servidores MCP, a resposta da Messages API informa os tokens usados em seu objeto `usage`.

<Note>
  A contagem de tokens é uma **estimativa**. Em alguns casos, o número real de tokens de entrada usados ao criar uma mensagem pode diferir ligeiramente.

  As contagens de tokens podem incluir tokens adicionados automaticamente pela Anthropic para otimizações do sistema. **Você não é cobrado pelos tokens adicionados pelo sistema**. A cobrança reflete apenas o seu conteúdo.
</Note>

### Modelos suportados

Todos os [modelos ativos](https://platform.claude.com/docs/pt-BR/models/overview) suportam contagem de tokens.

<Note>
  Os modelos Claude 4.7 e posteriores e o Claude Mythos Preview usam um tokenizador mais recente. O mesmo texto de entrada produz aproximadamente 30 por cento mais tokens do que em modelos anteriores. O aumento exato depende do conteúdo e do formato da carga de trabalho. Reconte os prompts com o modelo que você planeja usar, em vez de reutilizar contagens medidas com modelos anteriores.
</Note>

### Contar tokens em mensagens básicas

<CodeGroup>
  ```bash cURL
  curl https://api.anthropic.com/v1/messages/count_tokens \
    -H "x-api-key: $ANTHROPIC_API_KEY" \
    -H "content-type: application/json" \
    -H "anthropic-version: 2023-06-01" \
    -d '{
      "model": "claude-opus-5-5",
      "system": "You are a scientist",
      "messages": [{
        "role": "user",
        "content": "Hello, Claude"
      }]
    }'
  ```

  ```bash CLI
  ant messages count-tokens \
    --model claude-opus-5-5 \
    --system "You are a scientist" \
    --message '{role: user, content: "Hello, Claude"}'
  ```

  ```python Python
  client = anthropic.Anthropic()

  response = client.messages.count_tokens(
      model="claude-opus-5-5",
      system="You are a scientist",
      messages=[{"role": "user", "content": "Hello, Claude"}],
  )

  print(response.json())
  ```

  ```typescript TypeScript
  const client = new Anthropic();

  const response = await client.messages.countTokens({
    model: "claude-opus-5-5",
    system: "You are a scientist",
    messages: [
      {
        role: "user",
        content: "Hello, Claude"
      }
    ]
  });

  console.log(response);
  ```

  ```csharp C#
  using System;
  using System.Threading.Tasks;
  using Anthropic;
  using Anthropic.Models.Messages;

  AnthropicClient client = new();

  var parameters = new MessageCountTokensParams
  {
      Model = Model.ClaudeOpus5_5,
      System = "You are a scientist",
      Messages = [new() { Role = Role.User, Content = "Hello, Claude" }]
  };

  var response = await client.Messages.CountTokens(parameters);
  Console.WriteLine(response);
  ```

  ```go Go
  client := anthropic.NewClient()

  response, err := client.Messages.CountTokens(context.TODO(), anthropic.MessageCountTokensParams{
  	Model: anthropic.ModelClaudeOpus5_5,
  	System: anthropic.MessageCountTokensParamsSystemUnion{
  		OfString: anthropic.String("You are a scientist"),
  	},
  	Messages: []anthropic.MessageParam{
  		anthropic.NewUserMessage(anthropic.NewTextBlock("Hello, Claude")),
  	},
  })
  if err != nil {
  	log.Fatal(err)
  }

  fmt.Println(response)
  ```

  ```java Java
  import com.anthropic.models.messages.MessageCountTokensParams;
  import com.anthropic.models.messages.MessageTokensCount;
  // ...

  public class CountTokensExample {

    public static void main(String[] args) {
      AnthropicClient client = AnthropicOkHttpClient.fromEnv();

      MessageCountTokensParams params = MessageCountTokensParams.builder()
        .model(Model.CLAUDE_OPUS_5_5)
        .system("You are a scientist")
        .addUserMessage("Hello, Claude")
        .build();

      MessageTokensCount count = client.messages().countTokens(params);
      System.out.println(count);
    }
  }
  ```

  ```php PHP
  $client = new Client();

  $response = $client->messages->countTokens(
      messages: [
          ['role' => 'user', 'content' => 'Hello, Claude']
      ],
      model: 'claude-opus-5-5',
      system: 'You are a scientist',
  );

  echo json_encode($response);
  ```

  ```ruby Ruby
  client = Anthropic::Client.new

  response = client.messages.count_tokens(
    model: "claude-opus-5-5",
    system: "You are a scientist",
    messages: [
      { role: "user", content: "Hello, Claude" }
    ]
  )

  puts response
  ```
</CodeGroup>

```json Output
{ "input_tokens": 14 }
```

### Contar tokens em mensagens com ferramentas

<Note>
  A contagem de tokens suporta ferramentas de cliente e a [ferramenta advisor](https://platform.claude.com/docs/pt-BR/agents-and-tools/tool-use/advisor-tool). Requisições que incluem outras [ferramentas de servidor](https://platform.claude.com/docs/pt-BR/agents-and-tools/tool-use/server-tools) retornam um erro. Para a ferramenta advisor, a contagem abrange apenas a primeira chamada de amostragem do executor.
</Note>

<CodeGroup>
  ```bash cURL
  curl https://api.anthropic.com/v1/messages/count_tokens \
    -H "x-api-key: $ANTHROPIC_API_KEY" \
    -H "content-type: application/json" \
    -H "anthropic-version: 2023-06-01" \
    -d '{
      "model": "claude-opus-5-5",
      "tools": [
        {
          "name": "get_weather",
          "description": "Get the current weather in a given location",
          "input_schema": {
            "type": "object",
            "properties": {
              "location": {
                "type": "string",
                "description": "The city and state, e.g. San Francisco, CA"
              }
            },
            "required": ["location"]
          }
        }
      ],
      "messages": [
        {
          "role": "user",
          "content": "What'\''s the weather like in San Francisco?"
        }
      ]
    }'
  ```

  ```bash CLI
  ant messages count-tokens <<'YAML'
  model: claude-opus-5-5
  tools:
    - name: get_weather
      description: Get the current weather in a given location
      input_schema:
        type: object
        properties:
          location:
            type: string
            description: The city and state, e.g. San Francisco, CA
        required:
          - location
  messages:
    - role: user
      content: What's the weather like in San Francisco?
  YAML
  ```

  ```python Python
  client = anthropic.Anthropic()

  response = client.messages.count_tokens(
      model="claude-opus-5-5",
      tools=[
          {
              "name": "get_weather",
              "description": "Get the current weather in a given location",
              "input_schema": {
                  "type": "object",
                  "properties": {
                      "location": {
                          "type": "string",
                          "description": "The city and state, e.g. San Francisco, CA",
                      }
                  },
                  "required": ["location"],
              },
          }
      ],
      messages=[{"role": "user", "content": "What's the weather like in San Francisco?"}],
  )

  print(response.json())
  ```

  ```typescript TypeScript
  const client = new Anthropic();

  const response = await client.messages.countTokens({
    model: "claude-opus-5-5",
    tools: [
      {
        name: "get_weather",
        description: "Get the current weather in a given location",
        input_schema: {
          type: "object",
          properties: {
            location: {
              type: "string",
              description: "The city and state, e.g. San Francisco, CA"
            }
          },
          required: ["location"]
        }
      }
    ],
    messages: [{ role: "user", content: "What's the weather like in San Francisco?" }]
  });

  console.log(response);
  ```

  ```csharp C#
  using System;
  using System.Collections.Generic;
  using System.Text.Json;
  using System.Threading.Tasks;
  using Anthropic;
  using Anthropic.Models.Messages;

  AnthropicClient client = new();

  var parameters = new MessageCountTokensParams
  {
      Model = Model.ClaudeOpus5_5,
      Tools =
      [
          new MessageCountTokensTool(new Tool()
          {
              Name = "get_weather",
              Description = "Get the current weather in a given location",
              InputSchema = new InputSchema()
              {
                  Properties = new Dictionary<string, JsonElement>
                  {
                      ["location"] = JsonSerializer.SerializeToElement(new { type = "string", description = "The city and state, e.g. San Francisco, CA" }),
                  },
                  Required = ["location"],
              },
          }),
      ],
      Messages = [new() { Role = Role.User, Content = "What's the weather like in San Francisco?" }]
  };

  var count = await client.Messages.CountTokens(parameters);
  Console.WriteLine(count);
  ```

  ```go Go
  client := anthropic.NewClient()

  response, err := client.Messages.CountTokens(context.TODO(), anthropic.MessageCountTokensParams{
  	Model: anthropic.ModelClaudeOpus5_5,
  	Tools: []anthropic.MessageCountTokensToolUnionParam{
  		{OfTool: &anthropic.ToolParam{
  			Name:        "get_weather",
  			Description: anthropic.String("Get the current weather in a given location"),
  			InputSchema: anthropic.ToolInputSchemaParam{
  				Properties: map[string]any{
  					"location": map[string]any{
  						"type":        "string",
  						"description": "The city and state, e.g. San Francisco, CA",
  					},
  				},
  				Required: []string{"location"},
  			},
  		}},
  	},
  	Messages: []anthropic.MessageParam{
  		anthropic.NewUserMessage(anthropic.NewTextBlock("What's the weather like in San Francisco?")),
  	},
  })
  if err != nil {
  	log.Fatal(err)
  }

  jsonData, _ := json.MarshalIndent(response, "", "  ")
  fmt.Println(string(jsonData))
  ```

  ```java Java
  import com.anthropic.models.messages.MessageCountTokensParams;
  import com.anthropic.models.messages.MessageTokensCount;
  // ...
      AnthropicClient client = AnthropicOkHttpClient.fromEnv();

      InputSchema schema = InputSchema.builder()
        .properties(
          JsonValue.from(
            Map.of(
              "location",
              Map.of(
                "type",
                "string",
                "description",
                "The city and state, e.g. San Francisco, CA"
              )
            )
          )
        )
        .putAdditionalProperty("required", JsonValue.from(List.of("location")))
        .build();

      MessageCountTokensParams params = MessageCountTokensParams.builder()
        .model(Model.CLAUDE_OPUS_5_5)
        .addTool(
          Tool.builder()
            .name("get_weather")
            .description("Get the current weather in a given location")
            .inputSchema(schema)
            .build()
        )
        .addUserMessage("What's the weather like in San Francisco?")
        .build();

      MessageTokensCount count = client.messages().countTokens(params);
      System.out.println(count);
  ```

  ```php PHP
  $client = new Client();

  $response = $client->messages->countTokens(
      messages: [
          ['role' => 'user', 'content' => "What's the weather like in San Francisco?"]
      ],
      model: 'claude-opus-5-5',
      tools: [
          [
              'name' => 'get_weather',
              'description' => 'Get the current weather in a given location',
              'input_schema' => [
                  'type' => 'object',
                  'properties' => [
                      'location' => [
                          'type' => 'string',
                          'description' => 'The city and state, e.g. San Francisco, CA'
                      ]
                  ],
                  'required' => ['location']
              ]
          ]
      ],
  );

  echo json_encode($response, JSON_PRETTY_PRINT);
  ```

  ```ruby Ruby
  client = Anthropic::Client.new

  response = client.messages.count_tokens(
    model: "claude-opus-5-5",
    tools: [
      {
        name: "get_weather",
        description: "Get the current weather in a given location",
        input_schema: {
          type: "object",
          properties: {
            location: {
              type: "string",
              description: "The city and state, e.g. San Francisco, CA"
            }
          },
          required: ["location"]
        }
      }
    ],
    messages: [
      { role: "user", content: "What's the weather like in San Francisco?" }
    ]
  )

  puts response
  ```
</CodeGroup>

```json Output
{ "input_tokens": 403 }
```

### Contar tokens em mensagens com imagens

<CodeGroup>
  ```bash cURL
  #!/bin/sh

  IMAGE_URL="https://platform.claude.com/docs/images/vision-example.jpg"
  IMAGE_MEDIA_TYPE="image/jpeg"
  IMAGE_BASE64=$(curl -s "$IMAGE_URL" | base64 | tr -d '\n')

  curl https://api.anthropic.com/v1/messages/count_tokens \
    -H "x-api-key: $ANTHROPIC_API_KEY" \
    -H "anthropic-version: 2023-06-01" \
    -H "content-type: application/json" \
    -d @- <<EOF
  {
    "model": "claude-opus-5-5",
    "messages": [
      {"role": "user", "content": [
        {"type": "image", "source": {
          "type": "base64",
          "media_type": "$IMAGE_MEDIA_TYPE",
          "data": "$IMAGE_BASE64"
        }},
        {"type": "text", "text": "Describe this image"}
      ]}
    ]
  }
  EOF
  ```

  ```bash CLI
  IMAGE_URL="https://platform.claude.com/docs/images/vision-example.jpg"
  curl -s "$IMAGE_URL" -o ./vision-example.jpg

  ant messages count-tokens <<'YAML'
  model: claude-opus-5-5
  messages:
    - role: user
      content:
        - type: image
          source:
            type: base64
            media_type: image/jpeg
            data: "@./vision-example.jpg"
        - type: text
          text: Describe this image
  YAML
  ```

  ```python Python
  import base64
  import httpx2

  image_url = "https://platform.claude.com/docs/images/vision-example.jpg"
  image_media_type = "image/jpeg"
  image_data = base64.standard_b64encode(httpx2.get(image_url).content).decode("utf-8")

  client = anthropic.Anthropic()

  response = client.messages.count_tokens(
      model="claude-opus-5-5",
      messages=[
          {
              "role": "user",
              "content": [
                  {
                      "type": "image",
                      "source": {
                          "type": "base64",
                          "media_type": image_media_type,
                          "data": image_data,
                      },
                  },
                  {"type": "text", "text": "Describe this image"},
              ],
          }
      ],
  )
  print(response.json())
  ```

  ```typescript TypeScript
  const anthropic = new Anthropic();

  const imageUrl = "https://platform.claude.com/docs/images/vision-example.jpg";
  const imageMediaType = "image/jpeg";
  const imageArrayBuffer = await (await fetch(imageUrl)).arrayBuffer();
  const imageData = Buffer.from(imageArrayBuffer).toString("base64");

  const response = await anthropic.messages.countTokens({
    model: "claude-opus-5-5",
    messages: [
      {
        role: "user",
        content: [
          {
            type: "image",
            source: {
              type: "base64",
              media_type: imageMediaType,
              data: imageData
            }
          },
          {
            type: "text",
            text: "Describe this image"
          }
        ]
      }
    ]
  });
  console.log(response);
  ```

  ```csharp C#
  using System;
  using System.Collections.Generic;
  using System.Net.Http;
  using System.Threading.Tasks;
  using Anthropic;
  using Anthropic.Models.Messages;

  AnthropicClient client = new();

  string imageUrl = "https://platform.claude.com/docs/images/vision-example.jpg";

  using HttpClient httpClient = new();
  byte[] imageBytes = await httpClient.GetByteArrayAsync(imageUrl);
  string imageData = Convert.ToBase64String(imageBytes);

  var parameters = new MessageCountTokensParams
  {
      Model = Model.ClaudeOpus5_5,
      Messages =
      [
          new()
          {
              Role = Role.User,
              Content = new MessageParamContent(new List<ContentBlockParam>
              {
                  new ContentBlockParam(new ImageBlockParam(
                      new ImageBlockParamSource(new Base64ImageSource()
                      {
                          Data = imageData,
                          MediaType = MediaType.ImageJpeg,
                      })
                  )),
                  new ContentBlockParam(new TextBlockParam("Describe this image")),
              }),
          }
      ]
  };

  var count = await client.Messages.CountTokens(parameters);
  Console.WriteLine(count);
  ```

  ```go Go
  imageURL := "https://platform.claude.com/docs/images/vision-example.jpg"

  req, err := http.NewRequest("GET", imageURL, nil)
  if err != nil {
  	log.Fatal(err)
  }
  req.Header.Set("User-Agent", "AnthropicDocsBot/1.0")

  resp, err := http.DefaultClient.Do(req)
  if err != nil {
  	log.Fatal(err)
  }
  defer resp.Body.Close()

  imageBytes, err := io.ReadAll(resp.Body)
  if err != nil {
  	log.Fatal(err)
  }
  imageData := base64.StdEncoding.EncodeToString(imageBytes)

  client := anthropic.NewClient()

  response, err := client.Messages.CountTokens(context.TODO(), anthropic.MessageCountTokensParams{
  	Model: anthropic.ModelClaudeOpus5_5,
  	Messages: []anthropic.MessageParam{
  		anthropic.NewUserMessage(
  			anthropic.NewImageBlockBase64("image/jpeg", imageData),
  			anthropic.NewTextBlock("Describe this image"),
  		),
  	},
  })
  if err != nil {
  	log.Fatal(err)
  }

  fmt.Println(response)
  ```

  ```java Java
  import com.anthropic.models.messages.Base64ImageSource;
  // ...
  import com.anthropic.models.messages.MessageCountTokensParams;
  import com.anthropic.models.messages.MessageTokensCount;
  // ...
      AnthropicClient client = AnthropicOkHttpClient.fromEnv();

      String imageUrl =
        "https://platform.claude.com/docs/images/vision-example.jpg";
      String imageMediaType = "image/jpeg";

      HttpClient httpClient = HttpClient.newHttpClient();
      HttpRequest request = HttpRequest.newBuilder().uri(URI.create(imageUrl)).build();
      byte[] imageBytes = httpClient
        .send(request, HttpResponse.BodyHandlers.ofByteArray())
        .body();
      String imageBase64 = Base64.getEncoder().encodeToString(imageBytes);

      ContentBlockParam imageBlock = ContentBlockParam.ofImage(
        ImageBlockParam.builder()
          .source(
            Base64ImageSource.builder()
              .mediaType(Base64ImageSource.MediaType.IMAGE_JPEG)
              .data(imageBase64)
              .build()
          )
          .build()
      );

      ContentBlockParam textBlock = ContentBlockParam.ofText(
        TextBlockParam.builder().text("Describe this image").build()
      );

      MessageCountTokensParams params = MessageCountTokensParams.builder()
        .model(Model.CLAUDE_OPUS_5_5)
        .addUserMessageOfBlockParams(List.of(imageBlock, textBlock))
        .build();

      MessageTokensCount count = client.messages().countTokens(params);
      System.out.println(count);
  ```

  ```php PHP
  $imageUrl = "https://platform.claude.com/docs/images/vision-example.jpg";
  $imageMediaType = "image/jpeg";
  $imageData = base64_encode(file_get_contents($imageUrl));

  $client = new Client();

  $response = $client->messages->countTokens(
      messages: [
          [
              'role' => 'user',
              'content' => [
                  [
                      'type' => 'image',
                      'source' => [
                          'type' => 'base64',
                          'media_type' => $imageMediaType,
                          'data' => $imageData
                      ]
                  ],
                  ['type' => 'text', 'text' => 'Describe this image']
              ]
          ]
      ],
      model: 'claude-opus-5-5',
  );
  print_r($response);
  ```

  ```ruby Ruby
  require "base64"
  require "net/http"

  image_url = "https://platform.claude.com/docs/images/vision-example.jpg"
  image_media_type = "image/jpeg"

  uri = URI(image_url)
  image_data = Base64.strict_encode64(Net::HTTP.get(uri))

  client = Anthropic::Client.new

  response = client.messages.count_tokens(
    model: "claude-opus-5-5",
    messages: [
      {
        role: "user",
        content: [
          {
            type: "image",
            source: {
              type: "base64",
              media_type: image_media_type,
              data: image_data
            }
          },
          { type: "text", text: "Describe this image" }
        ]
      }
    ]
  )
  puts response
  ```
</CodeGroup>

```json Output
{ "input_tokens": 1028 }
```

Um bloco de imagem incorporado que define [`"oversized_image": "error"`](https://platform.claude.com/docs/pt-BR/build-with-claude/vision-coordinates#oversized-image-error) é rejeitado no momento da contagem exatamente como a Messages API o rejeitaria.

### Contar tokens em mensagens com pensamento

<Note>
  Consulte [Pensamento e a janela de contexto](https://platform.claude.com/docs/pt-BR/build-with-claude/thinking#thinking-and-the-context-window) para mais detalhes.

  * Blocos de pensamento de turnos **anteriores** do assistente contam para seus tokens de entrada em modelos que [mantêm todos os turnos anteriores](https://platform.claude.com/docs/pt-BR/build-with-claude/thinking#thinking-block-preservation-by-model); em modelos que mantêm apenas o último turno, a API os remove e eles **não** contam
  * O pensamento do turno **atual** do assistente **conta** para seus tokens de entrada
</Note>

<CodeGroup>
  ```bash cURL
  curl https://api.anthropic.com/v1/messages/count_tokens \
    -H "x-api-key: $ANTHROPIC_API_KEY" \
    -H "content-type: application/json" \
    -H "anthropic-version: 2023-06-01" \
    -d '{
      "model": "claude-opus-5-5",
      "thinking": {
        "type": "adaptive"
      },
      "messages": [
        {
          "role": "user",
          "content": "Are there an infinite number of prime numbers such that n mod 4 == 3?"
        },
        {
          "role": "assistant",
          "content": [
            {
              "type": "thinking",
              "thinking": "This is a nice number theory question. Lets think about it step by step...",
              "signature": "EuYBCkQYAiJAgCs1le6/Pol5Z4/JMomVOouGrWdhYNsH3ukzUECbB6iWrSQtsQuRHJID6lWV..."
            },
            {
              "type": "text",
              "text": "Yes, there are infinitely many prime numbers p such that p mod 4 = 3..."
            }
          ]
        },
        {
          "role": "user",
          "content": "Can you write a formal proof?"
        }
      ]
    }'
  ```

  ```bash CLI
  ant messages count-tokens <<'YAML'
  model: claude-opus-5-5
  thinking:
    type: adaptive
  messages:
    - role: user
      content: Are there an infinite number of prime numbers such that n mod 4 == 3?
    - role: assistant
      content:
        - type: thinking
          thinking: >-
            This is a nice number theory question. Lets think about it step by step...
          signature: >-
            EuYBCkQYAiJAgCs1le6/Pol5Z4/JMomVOouGrWdhYNsH3ukzUECbB6iWrSQtsQuRHJID6lWV...
        - type: text
          text: Yes, there are infinitely many prime numbers p such that p mod 4 = 3...
    - role: user
      content: Can you write a formal proof?
  YAML
  ```

  ```python Python
  client = anthropic.Anthropic()

  response = client.messages.count_tokens(
      model="claude-opus-5-5",
      thinking={"type": "adaptive"},
      messages=[
          {
              "role": "user",
              "content": "Are there an infinite number of prime numbers such that n mod 4 == 3?",
          },
          {
              "role": "assistant",
              "content": [
                  {
                      "type": "thinking",
                      "thinking": "This is a nice number theory question. Let's think about it step by step...",
                      "signature": "EuYBCkQYAiJAgCs1le6/Pol5Z4/JMomVOouGrWdhYNsH3ukzUECbB6iWrSQtsQuRHJID6lWV...",
                  },
                  {
                      "type": "text",
                      "text": "Yes, there are infinitely many prime numbers p such that p mod 4 = 3...",
                  },
              ],
          },
          {"role": "user", "content": "Can you write a formal proof?"},
      ],
  )

  print(response.json())
  ```

  ```typescript TypeScript
  const client = new Anthropic();

  const response = await client.messages.countTokens({
    model: "claude-opus-5-5",
    thinking: { type: "adaptive" },
    messages: [
      {
        role: "user",
        content: "Are there an infinite number of prime numbers such that n mod 4 == 3?"
      },
      {
        role: "assistant",
        content: [
          {
            type: "thinking",
            thinking:
              "This is a nice number theory question. Let's think about it step by step...",
            signature:
              "EuYBCkQYAiJAgCs1le6/Pol5Z4/JMomVOouGrWdhYNsH3ukzUECbB6iWrSQtsQuRHJID6lWV..."
          },
          {
            type: "text",
            text: "Yes, there are infinitely many prime numbers p such that p mod 4 = 3..."
          }
        ]
      },
      {
        role: "user",
        content: "Can you write a formal proof?"
      }
    ]
  });

  console.log(response);
  ```

  ```csharp C#
  using System;
  using System.Threading.Tasks;
  using System.Collections.Generic;
  using Anthropic;
  using Anthropic.Models.Messages;

  AnthropicClient client = new();

  var parameters = new MessageCountTokensParams
  {
      Model = Model.ClaudeOpus5_5,
      Thinking = new ThinkingConfigAdaptive(),
      Messages =
      [
          new()
          {
              Role = Role.User,
              Content = "Are there an infinite number of prime numbers such that n mod 4 == 3?"
          },
          new()
          {
              Role = Role.Assistant,
              Content = new MessageParamContent(new List<ContentBlockParam>
              {
                  new ContentBlockParam(new ThinkingBlockParam()
                  {
                      Thinking = "This is a nice number theory question. Let's think about it step by step...",
                      Signature = "EuYBCkQYAiJAgCs1le6/Pol5Z4/JMomVOouGrWdhYNsH3ukzUECbB6iWrSQtsQuRHJID6lWV...",
                  }),
                  new ContentBlockParam(new TextBlockParam("Yes, there are infinitely many prime numbers p such that p mod 4 = 3...")),
              }),
          },
          new()
          {
              Role = Role.User,
              Content = "Can you write a formal proof?"
          }
      ]
  };

  var response = await client.Messages.CountTokens(parameters);
  Console.WriteLine(response);
  ```

  ```go Go
  client := anthropic.NewClient()

  thinkingBlock := anthropic.NewThinkingBlock(
  	"EuYBCkQYAiJAgCs1le6/Pol5Z4/JMomVOouGrWdhYNsH3ukzUECbB6iWrSQtsQuRHJID6lWV...",
  	"This is a nice number theory question. Let's think about it step by step...",
  )

  textBlock := anthropic.NewTextBlock(
  	"Yes, there are infinitely many prime numbers p such that p mod 4 = 3...",
  )

  response, err := client.Messages.CountTokens(context.TODO(), anthropic.MessageCountTokensParams{
  	Model: anthropic.ModelClaudeOpus5_5,
  	Thinking: anthropic.ThinkingConfigParamUnion{
  		OfAdaptive: &anthropic.ThinkingConfigAdaptiveParam{},
  	},
  	Messages: []anthropic.MessageParam{
  		anthropic.NewUserMessage(anthropic.NewTextBlock("Are there an infinite number of prime numbers such that n mod 4 == 3?")),
  		anthropic.NewAssistantMessage(thinkingBlock, textBlock),
  		anthropic.NewUserMessage(anthropic.NewTextBlock("Can you write a formal proof?")),
  	},
  })
  if err != nil {
  	log.Fatal(err)
  }

  fmt.Printf("%+v\n", response)
  ```

  ```java Java
  import com.anthropic.models.messages.MessageCountTokensParams;
  import com.anthropic.models.messages.MessageTokensCount;
  // ...
  import com.anthropic.models.messages.ThinkingBlockParam;
  import com.anthropic.models.messages.ThinkingConfigAdaptive;
  // ...
      AnthropicClient client = AnthropicOkHttpClient.fromEnv();

      List<ContentBlockParam> assistantBlocks = List.of(
        ContentBlockParam.ofThinking(
          ThinkingBlockParam.builder()
            .thinking(
              "This is a nice number theory question. Let's think about it step by step..."
            )
            .signature(
              "EuYBCkQYAiJAgCs1le6/Pol5Z4/JMomVOouGrWdhYNsH3ukzUECbB6iWrSQtsQuRHJID6lWV..."
            )
            .build()
        ),
        ContentBlockParam.ofText(
          TextBlockParam.builder()
            .text("Yes, there are infinitely many prime numbers p such that p mod 4 = 3...")
            .build()
        )
      );

      MessageCountTokensParams params = MessageCountTokensParams.builder()
        .model(Model.CLAUDE_OPUS_5_5)
        .thinking(ThinkingConfigAdaptive.builder().build())
        .addUserMessage("Are there an infinite number of prime numbers such that n mod 4 == 3?")
        .addAssistantMessageOfBlockParams(assistantBlocks)
        .addUserMessage("Can you write a formal proof?")
        .build();

      MessageTokensCount count = client.messages().countTokens(params);
      System.out.println(count);
  ```

  ```php PHP
  $client = new Client();

  $response = $client->messages->countTokens(
      messages: [
          [
              'role' => 'user',
              'content' => 'Are there an infinite number of prime numbers such that n mod 4 == 3?'
          ],
          [
              'role' => 'assistant',
              'content' => [
                  [
                      'type' => 'thinking',
                      'thinking' => 'This is a nice number theory question. Let\'s think about it step by step...',
                      'signature' => 'EuYBCkQYAiJAgCs1le6/Pol5Z4/JMomVOouGrWdhYNsH3ukzUECbB6iWrSQtsQuRHJID6lWV...'
                  ],
                  [
                      'type' => 'text',
                      'text' => 'Yes, there are infinitely many prime numbers p such that p mod 4 = 3...'
                  ]
              ]
          ],
          [
              'role' => 'user',
              'content' => 'Can you write a formal proof?'
          ]
      ],
      model: 'claude-opus-5-5',
      thinking: ['type' => 'adaptive'],
  );

  echo json_encode($response);
  ```

  ```ruby Ruby
  client = Anthropic::Client.new

  response = client.messages.count_tokens(
    model: "claude-opus-5-5",
    thinking: {
      type: "adaptive"
    },
    messages: [
      {
        role: "user",
        content: "Are there an infinite number of prime numbers such that n mod 4 == 3?"
      },
      {
        role: "assistant",
        content: [
          {
            type: "thinking",
            thinking: "This is a nice number theory question. Let's think about it step by step...",
            signature: "EuYBCkQYAiJAgCs1le6/Pol5Z4/JMomVOouGrWdhYNsH3ukzUECbB6iWrSQtsQuRHJID6lWV..."
          },
          {
            type: "text",
            text: "Yes, there are infinitely many prime numbers p such that p mod 4 = 3..."
          }
        ]
      },
      {
        role: "user",
        content: "Can you write a formal proof?"
      }
    ]
  )

  puts response
  ```
</CodeGroup>

```json Output
{ "input_tokens": 88 }
```

### Contar tokens em mensagens com PDFs

<Note>
  A contagem de tokens suporta PDFs codificados em base64 com os mesmos [requisitos de PDF](https://platform.claude.com/docs/pt-BR/build-with-claude/pdf-support#check-pdf-requirements) da Messages API. Este endpoint não suporta origens de documento `url` ou `file`.
</Note>

<CodeGroup>
  ```bash cURL
  curl https://api.anthropic.com/v1/messages/count_tokens \
    -H "x-api-key: $ANTHROPIC_API_KEY" \
    -H "content-type: application/json" \
    -H "anthropic-version: 2023-06-01" \
    -d @- <<EOF
  {
    "model": "claude-opus-5-5",
    "messages": [{
      "role": "user",
      "content": [
        {
          "type": "document",
          "source": {
            "type": "base64",
            "media_type": "application/pdf",
            "data": "$PDF_BASE64"
          }
        },
        {
          "type": "text",
          "text": "Please summarize this document."
        }
      ]
    }]
  }
  EOF
  ```

  ```bash CLI
  ant messages count-tokens <<'YAML'
  model: claude-opus-5-5
  messages:
    - role: user
      content:
        - type: document
          source:
            type: base64
            media_type: application/pdf
            data: "@./document.pdf"
        - type: text
          text: Please summarize this document.
  YAML
  ```

  ```python Python
  import base64
  import anthropic

  client = anthropic.Anthropic()

  with open("/path/to/document.pdf", "rb") as pdf_file:
      pdf_base64 = base64.standard_b64encode(pdf_file.read()).decode("utf-8")

  response = client.messages.count_tokens(
      model="claude-opus-5-5",
      messages=[
          {
              "role": "user",
              "content": [
                  {
                      "type": "document",
                      "source": {
                          "type": "base64",
                          "media_type": "application/pdf",
                          "data": pdf_base64,
                      },
                  },
                  {"type": "text", "text": "Please summarize this document."},
              ],
          }
      ],
  )

  print(response.json())
  ```

  ```typescript TypeScript
  import { readFile } from "node:fs/promises";

  const client = new Anthropic();

  const pdfBase64 = await readFile("/path/to/document.pdf", { encoding: "base64" });

  const response = await client.messages.countTokens({
    model: "claude-opus-5-5",
    messages: [
      {
        role: "user",
        content: [
          {
            type: "document",
            source: {
              type: "base64",
              media_type: "application/pdf",
              data: pdfBase64
            }
          },
          {
            type: "text",
            text: "Please summarize this document."
          }
        ]
      }
    ]
  });

  console.log(response);
  ```

  ```csharp C#
  using System;
  using System.IO;
  using System.Threading.Tasks;
  using System.Collections.Generic;
  using Anthropic;
  using Anthropic.Models.Messages;

  AnthropicClient client = new();

  byte[] pdfBytes = await File.ReadAllBytesAsync("/path/to/document.pdf");
  string pdfBase64 = Convert.ToBase64String(pdfBytes);

  var parameters = new MessageCountTokensParams
  {
      Model = Model.ClaudeOpus5_5,
      Messages =
      [
          new()
          {
              Role = Role.User,
              Content = new MessageParamContent(new List<ContentBlockParam>
              {
                  new ContentBlockParam(new DocumentBlockParam(
                      new DocumentBlockParamSource(new Base64PdfSource()
                      {
                          Data = pdfBase64,
                      })
                  )),
                  new ContentBlockParam(new TextBlockParam("Please summarize this document.")),
              }),
          }
      ]
  };

  var count = await client.Messages.CountTokens(parameters);
  Console.WriteLine(count);
  ```

  ```go Go
  client := anthropic.NewClient()

  pdfBytes, err := os.ReadFile("/path/to/document.pdf")
  if err != nil {
  	log.Fatal(err)
  }
  pdfBase64 := base64.StdEncoding.EncodeToString(pdfBytes)

  response, err := client.Messages.CountTokens(context.TODO(), anthropic.MessageCountTokensParams{
  	Model: anthropic.ModelClaudeOpus5_5,
  	Messages: []anthropic.MessageParam{
  		anthropic.NewUserMessage(
  			anthropic.NewDocumentBlock(anthropic.Base64PDFSourceParam{
  				Data: pdfBase64,
  			}),
  			anthropic.NewTextBlock("Please summarize this document."),
  		),
  	},
  })
  if err != nil {
  	log.Fatal(err)
  }

  fmt.Println(response)
  ```

  ```java Java
  import com.anthropic.models.messages.Base64PdfSource;
  // ...
  import com.anthropic.models.messages.DocumentBlockParam;
  import com.anthropic.models.messages.MessageCountTokensParams;
  import com.anthropic.models.messages.MessageTokensCount;
  // ...
      AnthropicClient client = AnthropicOkHttpClient.fromEnv();

      byte[] fileBytes = Files.readAllBytes(Path.of("/path/to/document.pdf"));
      String pdfBase64 = Base64.getEncoder().encodeToString(fileBytes);

      ContentBlockParam documentBlock = ContentBlockParam.ofDocument(
        DocumentBlockParam.builder()
          .source(Base64PdfSource.builder().data(pdfBase64).build())
          .build()
      );

      ContentBlockParam textBlock = ContentBlockParam.ofText(
        TextBlockParam.builder().text("Please summarize this document.").build()
      );

      MessageCountTokensParams params = MessageCountTokensParams.builder()
        .model(Model.CLAUDE_OPUS_5_5)
        .addUserMessageOfBlockParams(List.of(documentBlock, textBlock))
        .build();

      MessageTokensCount count = client.messages().countTokens(params);
      System.out.println(count);
  ```

  ```php PHP
  $client = new Client();

  $pdfBase64 = base64_encode(file_get_contents("/path/to/document.pdf"));

  $response = $client->messages->countTokens(
      messages: [
          [
              'role' => 'user',
              'content' => [
                  [
                      'type' => 'document',
                      'source' => [
                          'type' => 'base64',
                          'media_type' => 'application/pdf',
                          'data' => $pdfBase64
                      ]
                  ],
                  [
                      'type' => 'text',
                      'text' => 'Please summarize this document.'
                  ]
              ]
          ]
      ],
      model: 'claude-opus-5-5',
  );

  echo json_encode($response);
  ```

  ```ruby Ruby
  require "base64"

  client = Anthropic::Client.new

  pdf_base64 = Base64.strict_encode64(File.binread("/path/to/document.pdf"))

  response = client.messages.count_tokens(
    model: "claude-opus-5-5",
    messages: [
      {
        role: "user",
        content: [
          {
            type: "document",
            source: {
              type: "base64",
              media_type: "application/pdf",
              data: pdf_base64
            }
          },
          {
            type: "text",
            text: "Please summarize this document."
          }
        ]
      }
    ]
  )

  puts response
  ```
</CodeGroup>

```json Output
{ "input_tokens": 2188 }
```

***

## Contagens de tokens nos modelos Claude Fable e Claude Mythos

Claude Fable 5.1, Claude Mythos 5.1, Claude Fable 5 e Claude Mythos 5 compartilham o tokenizador introduzido com o Claude Opus 4.7. Um prompt tem a mesma contagem nos quatro, e aproximadamente 30 por cento maior do que em modelos anteriores ao Claude Opus 4.7 (o aumento exato depende do conteúdo). O endpoint de contagem de tokens conta usando o tokenizador do `model` que você informar. Para medir a diferença para sua carga de trabalho, conte a mesma requisição duas vezes, uma com seu modelo atual e outra com o modelo para o qual você planeja migrar, e compare os dois valores de `input_tokens`.

<Note>
  **Cobrança e migração:** O uso e a cobrança nesses modelos refletem as contagens deste tokenizador. Ao migrar de um modelo anterior ao Claude Opus 4.7, não reutilize contagens de tokens medidas no modelo mais antigo para estimar custos ou o ajuste à "context window" (janela de contexto). Conte seus prompts com o ID de `model` que você planeja usar (por exemplo, `"claude-fable-5-1"`).
</Note>

***

## Preços e limites de taxa

A contagem de tokens é **gratuita**, mas está sujeita a limites de taxa de requisições por minuto com base no seu [nível de uso](https://platform.claude.com/docs/pt-BR/api/rate-limits#rate-limits). Se você precisar de limites mais altos, use **Request tier increase** na página [Rate limits](https://platform.claude.com/usage/limits).

| Nível de uso | Requisições por minuto (RPM) |
| ------------ | ---------------------------- |
| Start        | 5.000                        |
| Build        | 10.000                       |
| Scale        | 20.000                       |

<Note>
  A contagem de tokens e a criação de mensagens têm limites de taxa separados e independentes. O uso de um não conta para os limites do outro.
</Note>

***

## Perguntas frequentes

<AccordionGroup>
  <Accordion title="A contagem de tokens usa cache de prompt?">
    Não, a contagem de tokens fornece uma estimativa sem usar a lógica de cache. Embora você possa fornecer blocos `cache_control` em sua requisição de contagem de tokens, o cache de prompt ocorre apenas durante a criação real de mensagens.
  </Accordion>
</AccordionGroup>

***

## Próximos passos

<CardGroup cols={2}>
  <Card title="Contar tokens de mensagens" icon="code" href="https://platform.claude.com/docs/pt-BR/api/messages/count_tokens">
    Leia a referência completa da API para o endpoint de contagem de tokens.
  </Card>

  <Card title="Janelas de contexto" icon="arrows-maximize" href="https://platform.claude.com/docs/pt-BR/build-with-claude/context-windows">
    Use contagens de tokens para manter os prompts dentro da janela de contexto de um modelo.
  </Card>

  <Card title="Limites de taxa" icon="gauge" href="https://platform.claude.com/docs/pt-BR/api/rate-limits">
    Verifique as contagens de tokens antes de enviar uma requisição para permanecer dentro do seu nível de uso.
  </Card>

  <Card title="Cache de prompt" icon="database" href="https://platform.claude.com/docs/pt-BR/build-with-claude/prompt-caching">
    Reduza o custo e a latência em prompts repetidos armazenando prefixos de prompt em cache.
  </Card>
</CardGroup>