Count Tokens

预估一段 Anthropic Messages 请求会消耗的输入 token 数。不会调上游推理、不计费、不计入并发，但仍会校验余额或订阅状态以及 API Key 有效性。

协议边界

这是 Anthropic 风格的 token 预估接口，请求体与 /v1/messages 相同。Anthropic 平台分组转发上游计数，OpenAI 平台分组使用兼容计数，Grok 分组返回估算值。

Endpoint

POST https://api.sakrylle.com/v1/messages/count_tokens

请求头

名称	必填	说明
Authorization	是	Bearer sk-xxxxxxxxxxxxxxxx
Content-Type	是	application/json
anthropic-version	否	例如 2023-06-01，与 Anthropic 官方一致

请求体参数

请求体格式与 POST /v1/messages 完全一致：

字段	类型	必填	说明
model	string	是	Claude 模型 ID，例如 claude-haiku-4-5-20251001
messages	array	是	完整对话消息列表
system	string | array	否	系统提示
tools	array	否	工具定义；会一并计入输入 token
tool_choice	object	否	同 messages

所有用于推理的字段（temperature、top_p、stream、max_tokens 等）即便传了也只用于解析校验，不影响计数结果。

请求体示例

{
  "model": "claude-haiku-4-5-20251001",
  "system": "你是一个简洁的中文助手。",
  "messages": [
    { "role": "user", "content": "自我介绍一下。" }
  ]
}

响应示例

{
  "input_tokens": 36
}

字段命名与 Anthropic 官方 /v1/messages/count_tokens 保持一致；网关从被选中的上游账号回传该结果。

错误

• 400 invalid_request_error — 缺 model、请求体非 JSON 或解析失败
• 401 authentication_error — API Key 无效
• 403 billing_error — 余额或订阅限额不足（即使不计费也会校验）
• 503 api_error — 没有可用的上游账号

更多错误码见 错误处理。

代码示例

curl

curl https://api.sakrylle.com/v1/messages/count_tokens \
  -H "Authorization: Bearer sk-xxxxxxxxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -H "anthropic-version: 2023-06-01" \
  -d '{
    "model": "claude-haiku-4-5-20251001",
    "messages": [
      {"role": "user", "content": "你好"}
    ]
  }'

Python

from anthropic import Anthropic

client = Anthropic(
    api_key="sk-xxxxxxxxxxxxxxxx",
    base_url="https://api.sakrylle.com",
)

resp = client.messages.count_tokens(
    model="claude-haiku-4-5-20251001",
    messages=[{"role": "user", "content": "你好"}],
)
print(resp.input_tokens)

Node

import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic({
  apiKey: "sk-xxxxxxxxxxxxxxxx",
  baseURL: "https://api.sakrylle.com",
});

const resp = await client.messages.countTokens({
  model: "claude-haiku-4-5-20251001",
  messages: [{ role: "user", content: "你好" }],
});
console.log(resp.input_tokens);

Go

package main

import (
    "bytes"
    "fmt"
    "io"
    "net/http"
)

func main() {
    body := []byte(`{
      "model": "claude-haiku-4-5-20251001",
      "messages": [{"role":"user","content":"你好"}]
    }`)
    req, _ := http.NewRequest("POST",
        "https://api.sakrylle.com/v1/messages/count_tokens",
        bytes.NewReader(body))
    req.Header.Set("Authorization", "Bearer sk-xxxxxxxxxxxxxxxx")
    req.Header.Set("Content-Type", "application/json")
    req.Header.Set("anthropic-version", "2023-06-01")
    resp, err := http.DefaultClient.Do(req)
    if err != nil {
        panic(err)
    }
    defer resp.Body.Close()
    out, _ := io.ReadAll(resp.Body)
    fmt.Println(string(out))
}

Rust

use reqwest::Client;
use serde_json::json;

#[tokio::main]
async fn main() -> anyhow::Result<()> {
    let client = Client::new();
    let resp: serde_json::Value = client
        .post("https://api.sakrylle.com/v1/messages/count_tokens")
        .bearer_auth("sk-xxxxxxxxxxxxxxxx")
        .header("anthropic-version", "2023-06-01")
        .json(&json!({
            "model": "claude-haiku-4-5-20251001",
            "messages": [{"role": "user", "content": "你好"}]
        }))
        .send()
        .await?
        .json()
        .await?;
    println!("{resp:#}");
    Ok(())
}

Java

import java.net.URI;
import java.net.http.*;

public class CountTokens {
  public static void main(String[] args) throws Exception {
    String body = """
      {"model":"claude-haiku-4-5-20251001",
       "messages":[{"role":"user","content":"你好"}]}
      """;
    HttpRequest req = HttpRequest.newBuilder()
      .uri(URI.create("https://api.sakrylle.com/v1/messages/count_tokens"))
      .header("Authorization", "Bearer sk-xxxxxxxxxxxxxxxx")
      .header("Content-Type", "application/json")
      .header("anthropic-version", "2023-06-01")
      .POST(HttpRequest.BodyPublishers.ofString(body))
      .build();
    HttpResponse<String> resp = HttpClient.newHttpClient()
      .send(req, HttpResponse.BodyHandlers.ofString());
    System.out.println(resp.body());
  }
}

C#

using System.Net.Http.Headers;
using System.Text;

var http = new HttpClient();
http.DefaultRequestHeaders.Authorization =
    new AuthenticationHeaderValue("Bearer", "sk-xxxxxxxxxxxxxxxx");
http.DefaultRequestHeaders.Add("anthropic-version", "2023-06-01");

var payload = """
  {"model":"claude-haiku-4-5-20251001",
   "messages":[{"role":"user","content":"你好"}]}
  """;
var resp = await http.PostAsync(
    "https://api.sakrylle.com/v1/messages/count_tokens",
    new StringContent(payload, Encoding.UTF8, "application/json"));
Console.WriteLine(await resp.Content.ReadAsStringAsync());