Claude Messages API

curl https://api.apimart.ai/v1/messages \
  -H "x-api-key: $API_KEY" \
  -H "anthropic-version: 2025-10-01" \
  -H "content-type: application/json" \
  -d '{
    "model": "claude-sonnet-4-6",
    "max_tokens": 1024,
    "messages": [
      {"role": "user", "content": "Hello, world"}
    ]
  }'

import anthropic

client = anthropic.Anthropic(
    api_key="YOUR_API_KEY",
    base_url="https://api.apimart.ai"
)

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

print(message.content)

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

const client = new Anthropic({
  apiKey: process.env.API_KEY,
  baseURL: 'https://api.apimart.ai'
});

const message = await client.messages.create({
  model: 'claude-sonnet-4-6',
  max_tokens: 1024,
  messages: [
    { role: 'user', content: 'Hello, world' }
  ]
});

console.log(message.content);

package main

import (
    "bytes"
    "encoding/json"
    "fmt"
    "io/ioutil"
    "net/http"
    "os"
)

func main() {
    url := "https://api.apimart.ai/v1/messages"

    payload := map[string]interface{}{
        "model": "claude-sonnet-4-6",
        "max_tokens": 1024,
        "messages": []map[string]string{
            {
                "role":    "user",
                "content": "Hello, world",
            },
        },
    }

    jsonData, _ := json.Marshal(payload)

    req, _ := http.NewRequest("POST", url, bytes.NewBuffer(jsonData))
    req.Header.Set("x-api-key", os.Getenv("API_KEY"))
    req.Header.Set("anthropic-version", "2025-10-01")
    req.Header.Set("Content-Type", "application/json")

    client := &http.Client{}
    resp, err := client.Do(req)
    if err != nil {
        panic(err)
    }
    defer resp.Body.Close()

    body, _ := ioutil.ReadAll(resp.Body)
    fmt.Println(string(body))
}

import java.net.http.HttpClient;
import java.net.http.HttpRequest;
import java.net.http.HttpResponse;
import java.net.URI;

public class Main {
    public static void main(String[] args) throws Exception {
        String url = "https://api.apimart.ai/v1/messages";
        String apiKey = System.getenv("API_KEY");

        String payload = """
        {
          "model": "claude-sonnet-4-6",
          "max_tokens": 1024,
          "messages": [
            {
              "role": "user",
              "content": "Hello, world"
            }
          ]
        }
        """;

        HttpClient client = HttpClient.newHttpClient();
        HttpRequest request = HttpRequest.newBuilder()
            .uri(URI.create(url))
            .header("x-api-key", apiKey)
            .header("anthropic-version", "2025-10-01")
            .header("Content-Type", "application/json")
            .POST(HttpRequest.BodyPublishers.ofString(payload))
            .build();

        HttpResponse<String> response = client.send(request,
            HttpResponse.BodyHandlers.ofString());

        System.out.println(response.body());
    }
}

<?php

$url = "https://api.apimart.ai/v1/messages";
$apiKey = getenv('API_KEY');

$payload = [
    "model" => "claude-sonnet-4-6",
    "max_tokens" => 1024,
    "messages" => [
        [
            "role" => "user",
            "content" => "Hello, world"
        ]
    ]
];

$ch = curl_init($url);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
curl_setopt($ch, CURLOPT_POST, true);
curl_setopt($ch, CURLOPT_POSTFIELDS, json_encode($payload));
curl_setopt($ch, CURLOPT_HTTPHEADER, [
    "x-api-key: " . $apiKey,
    "anthropic-version: 2025-10-01",
    "Content-Type: application/json"
]);

$response = curl_exec($ch);
curl_close($ch);

echo $response;
?>

require 'net/http'
require 'json'
require 'uri'

url = URI("https://api.apimart.ai/v1/messages")
api_key = ENV['API_KEY']

payload = {
  model: "claude-sonnet-4-6",
  max_tokens: 1024,
  messages: [
    {
      role: "user",
      content: "Hello, world"
    }
  ]
}

http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true

request = Net::HTTP::Post.new(url)
request["x-api-key"] = api_key
request["anthropic-version"] = "2025-10-01"
request["Content-Type"] = "application/json"
request.body = payload.to_json

response = http.request(request)
puts response.body

import Foundation

let url = URL(string: "https://api.apimart.ai/v1/messages")!
let apiKey = ProcessInfo.processInfo.environment["API_KEY"] ?? ""

let payload: [String: Any] = [
    "model": "claude-sonnet-4-6",
    "max_tokens": 1024,
    "messages": [
        [
            "role": "user",
            "content": "Hello, world"
        ]
    ]
]

var request = URLRequest(url: url)
request.httpMethod = "POST"
request.setValue(apiKey, forHTTPHeaderField: "x-api-key")
request.setValue("2025-10-01", forHTTPHeaderField: "anthropic-version")
request.setValue("application/json", forHTTPHeaderField: "Content-Type")
request.httpBody = try? JSONSerialization.data(withJSONObject: payload)

let task = URLSession.shared.dataTask(with: request) { data, response, error in
    if let error = error {
        print("Error: \(error)")
        return
    }
    
    if let data = data, let responseString = String(data: data, encoding: .utf8) {
        print(responseString)
    }
}

task.resume()

using System;
using System.Net.Http;
using System.Text;
using System.Threading.Tasks;

class Program
{
    static async Task Main(string[] args)
    {
        var url = "https://api.apimart.ai/v1/messages";
        var apiKey = Environment.GetEnvironmentVariable("API_KEY");

        var payload = @"{
            ""model"": ""claude-sonnet-4-6"",
            ""max_tokens"": 1024,
            ""messages"": [
                {
                    ""role"": ""user"",
                    ""content"": ""Hello, world""
                }
            ]
        }";

        using var client = new HttpClient();
        client.DefaultRequestHeaders.Add("x-api-key", apiKey);
        client.DefaultRequestHeaders.Add("anthropic-version", "2025-10-01");

        var content = new StringContent(payload, Encoding.UTF8, "application/json");
        var response = await client.PostAsync(url, content);
        var result = await response.Content.ReadAsStringAsync();

        Console.WriteLine(result);
    }
}

#include <stdio.h>
#include <curl/curl.h>
#include <stdlib.h>

int main(void) {
    CURL *curl;
    CURLcode res;
    const char *api_key = getenv("API_KEY");

    curl_global_init(CURL_GLOBAL_DEFAULT);
    curl = curl_easy_init();

    if(curl) {
        const char *url = "https://api.apimart.ai/v1/messages";
        const char *payload = "{"
            "\"model\":\"claude-sonnet-4-6\","
            "\"max_tokens\":1024,"
            "\"messages\":[{\"role\":\"user\",\"content\":\"Hello, world\"}]"
        "}";

        char auth_header[256];
        snprintf(auth_header, sizeof(auth_header), "x-api-key: %s", api_key);

        struct curl_slist *headers = NULL;
        headers = curl_slist_append(headers, auth_header);
        headers = curl_slist_append(headers, "anthropic-version: 2025-10-01");
        headers = curl_slist_append(headers, "Content-Type: application/json");

        curl_easy_setopt(curl, CURLOPT_URL, url);
        curl_easy_setopt(curl, CURLOPT_POSTFIELDS, payload);
        curl_easy_setopt(curl, CURLOPT_HTTPHEADER, headers);

        res = curl_easy_perform(curl);

        if(res != CURLE_OK) {
            fprintf(stderr, "curl_easy_perform() failed: %s\n",
                    curl_easy_strerror(res));
        }

        curl_slist_free_all(headers);
        curl_easy_cleanup(curl);
    }

    curl_global_cleanup();
    return 0;
}

#import <Foundation/Foundation.h>

int main(int argc, const char * argv[]) {
    @autoreleasepool {
        NSURL *url = [NSURL URLWithString:@"https://api.apimart.ai/v1/messages"];
        NSString *apiKey = [NSProcessInfo processInfo].environment[@"API_KEY"];
        
        NSDictionary *payload = @{
            @"model": @"claude-sonnet-4-6",
            @"max_tokens": @1024,
            @"messages": @[
                @{
                    @"role": @"user",
                    @"content": @"Hello, world"
                }
            ]
        };
        
        NSError *error;
        NSData *jsonData = [NSJSONSerialization dataWithJSONObject:payload
                                                          options:0
                                                            error:&error];
        
        NSMutableURLRequest *request = [NSMutableURLRequest requestWithURL:url];
        [request setHTTPMethod:@"POST"];
        [request setValue:apiKey forHTTPHeaderField:@"x-api-key"];
        [request setValue:@"2025-10-01" forHTTPHeaderField:@"anthropic-version"];
        [request setValue:@"application/json" forHTTPHeaderField:@"Content-Type"];
        [request setHTTPBody:jsonData];
        
        NSURLSessionDataTask *task = [[NSURLSession sharedSession] 
            dataTaskWithRequest:request
            completionHandler:^(NSData *data, NSURLResponse *response, NSError *error) {
                if (error) {
                    NSLog(@"Error: %@", error);
                    return;
                }
                NSString *result = [[NSString alloc] initWithData:data 
                                                        encoding:NSUTF8StringEncoding];
                NSLog(@"%@", result);
            }];
        
        [task resume];
        [[NSRunLoop mainRunLoop] run];
    }
    return 0;
}

(* Requires cohttp and yojson libraries *)
open Lwt
open Cohttp
open Cohttp_lwt_unix

let url = "https://api.apimart.ai/v1/messages"
let api_key = Sys.getenv "API_KEY"

let payload = {|{
  "model": "claude-sonnet-4-6",
  "max_tokens": 1024,
  "messages": [
    {
      "role": "user",
      "content": "Hello, world"
    }
  ]
}|}

let () =
  let headers = Header.init ()
    |> fun h -> Header.add h "x-api-key" api_key
    |> fun h -> Header.add h "anthropic-version" "2025-10-01"
    |> fun h -> Header.add h "Content-Type" "application/json"
  in
  let body = Cohttp_lwt.Body.of_string payload in
  
  let response = Client.post ~headers ~body (Uri.of_string url) >>= fun (resp, body) ->
    body |> Cohttp_lwt.Body.to_string >|= fun body_str ->
    print_endline body_str
  in
  Lwt_main.run response

import 'dart:convert';
import 'dart:io';
import 'package:http/http.dart' as http;

void main() async {
  final url = Uri.parse('https://api.apimart.ai/v1/messages');
  final apiKey = Platform.environment['API_KEY'];
  
  final payload = {
    'model': 'claude-sonnet-4-6',
    'max_tokens': 1024,
    'messages': [
      {
        'role': 'user',
        'content': 'Hello, world'
      }
    ]
  };
  
  final response = await http.post(
    url,
    headers: {
      'x-api-key': apiKey!,
      'anthropic-version': '2025-10-01',
      'Content-Type': 'application/json',
    },
    body: jsonEncode(payload),
  );
  
  print(response.body);
}

library(httr)
library(jsonlite)

url <- "https://api.apimart.ai/v1/messages"
api_key <- Sys.getenv("API_KEY")

payload <- list(
  model = "claude-sonnet-4-6",
  max_tokens = 1024,
  messages = list(
    list(
      role = "user",
      content = "Hello, world"
    )
  )
)

response <- POST(
  url,
  add_headers(
    `x-api-key` = api_key,
    `anthropic-version` = "2025-10-01",
    `Content-Type` = "application/json"
  ),
  body = toJSON(payload, auto_unbox = TRUE),
  encode = "raw"
)

cat(content(response, "text"))

{
  "model": "claude-sonnet-4-6",
  "id": "msg_011CdfeHuC728oxqaLrRNbcB",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "type": "text",
      "text": "Hello! I'm Claude. Nice to meet you."
    }
  ],
  "stop_reason": "end_turn",
  "stop_sequence": null,
  "stop_details": null,
  "usage": {
    "input_tokens": 12,
    "cache_creation_input_tokens": 0,
    "cache_read_input_tokens": 0,
    "cache_creation": {
      "ephemeral_5m_input_tokens": 0,
      "ephemeral_1h_input_tokens": 0
    },
    "output_tokens": 18,
    "service_tier": "standard",
    "inference_geo": "global"
  }
}

{
  "error": {
    "code": "model_not_found",
    "message": "model not found (request id: 20260803181903172761862QzohKPXF)",
    "param": "",
    "type": "apimart_error"
  }
}

{
  "error": {
    "code": "",
    "message": "Invalid API key (request id: 20260803181903172761862QzohKPXF)",
    "param": "",
    "type": "apimart_error"
  }
}

{
  "error": {
    "code": "",
    "message": "Insufficient balance (request id: 20260803181903172761862QzohKPXF)",
    "param": "",
    "type": "apimart_error"
  }
}

{
  "error": {
    "code": "",
    "message": "Too many requests (request id: 20260803181903172761862QzohKPXF)",
    "param": "",
    "type": "apimart_error"
  }
}

{
  "error": {
    "code": "",
    "message": "Internal server error (request id: 20260803181903172761862QzohKPXF)",
    "param": "",
    "type": "apimart_error"
  }
}

POST

messages

curl https://api.apimart.ai/v1/messages \
  -H "x-api-key: $API_KEY" \
  -H "anthropic-version: 2025-10-01" \
  -H "content-type: application/json" \
  -d '{
    "model": "claude-sonnet-4-6",
    "max_tokens": 1024,
    "messages": [
      {"role": "user", "content": "Hello, world"}
    ]
  }'

import anthropic

client = anthropic.Anthropic(
    api_key="YOUR_API_KEY",
    base_url="https://api.apimart.ai"
)

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

print(message.content)

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

const client = new Anthropic({
  apiKey: process.env.API_KEY,
  baseURL: 'https://api.apimart.ai'
});

const message = await client.messages.create({
  model: 'claude-sonnet-4-6',
  max_tokens: 1024,
  messages: [
    { role: 'user', content: 'Hello, world' }
  ]
});

console.log(message.content);

package main

import (
    "bytes"
    "encoding/json"
    "fmt"
    "io/ioutil"
    "net/http"
    "os"
)

func main() {
    url := "https://api.apimart.ai/v1/messages"

    payload := map[string]interface{}{
        "model": "claude-sonnet-4-6",
        "max_tokens": 1024,
        "messages": []map[string]string{
            {
                "role":    "user",
                "content": "Hello, world",
            },
        },
    }

    jsonData, _ := json.Marshal(payload)

    req, _ := http.NewRequest("POST", url, bytes.NewBuffer(jsonData))
    req.Header.Set("x-api-key", os.Getenv("API_KEY"))
    req.Header.Set("anthropic-version", "2025-10-01")
    req.Header.Set("Content-Type", "application/json")

    client := &http.Client{}
    resp, err := client.Do(req)
    if err != nil {
        panic(err)
    }
    defer resp.Body.Close()

    body, _ := ioutil.ReadAll(resp.Body)
    fmt.Println(string(body))
}

import java.net.http.HttpClient;
import java.net.http.HttpRequest;
import java.net.http.HttpResponse;
import java.net.URI;

public class Main {
    public static void main(String[] args) throws Exception {
        String url = "https://api.apimart.ai/v1/messages";
        String apiKey = System.getenv("API_KEY");

        String payload = """
        {
          "model": "claude-sonnet-4-6",
          "max_tokens": 1024,
          "messages": [
            {
              "role": "user",
              "content": "Hello, world"
            }
          ]
        }
        """;

        HttpClient client = HttpClient.newHttpClient();
        HttpRequest request = HttpRequest.newBuilder()
            .uri(URI.create(url))
            .header("x-api-key", apiKey)
            .header("anthropic-version", "2025-10-01")
            .header("Content-Type", "application/json")
            .POST(HttpRequest.BodyPublishers.ofString(payload))
            .build();

        HttpResponse<String> response = client.send(request,
            HttpResponse.BodyHandlers.ofString());

        System.out.println(response.body());
    }
}

<?php

$url = "https://api.apimart.ai/v1/messages";
$apiKey = getenv('API_KEY');

$payload = [
    "model" => "claude-sonnet-4-6",
    "max_tokens" => 1024,
    "messages" => [
        [
            "role" => "user",
            "content" => "Hello, world"
        ]
    ]
];

$ch = curl_init($url);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
curl_setopt($ch, CURLOPT_POST, true);
curl_setopt($ch, CURLOPT_POSTFIELDS, json_encode($payload));
curl_setopt($ch, CURLOPT_HTTPHEADER, [
    "x-api-key: " . $apiKey,
    "anthropic-version: 2025-10-01",
    "Content-Type: application/json"
]);

$response = curl_exec($ch);
curl_close($ch);

echo $response;
?>

require 'net/http'
require 'json'
require 'uri'

url = URI("https://api.apimart.ai/v1/messages")
api_key = ENV['API_KEY']

payload = {
  model: "claude-sonnet-4-6",
  max_tokens: 1024,
  messages: [
    {
      role: "user",
      content: "Hello, world"
    }
  ]
}

http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true

request = Net::HTTP::Post.new(url)
request["x-api-key"] = api_key
request["anthropic-version"] = "2025-10-01"
request["Content-Type"] = "application/json"
request.body = payload.to_json

response = http.request(request)
puts response.body

import Foundation

let url = URL(string: "https://api.apimart.ai/v1/messages")!
let apiKey = ProcessInfo.processInfo.environment["API_KEY"] ?? ""

let payload: [String: Any] = [
    "model": "claude-sonnet-4-6",
    "max_tokens": 1024,
    "messages": [
        [
            "role": "user",
            "content": "Hello, world"
        ]
    ]
]

var request = URLRequest(url: url)
request.httpMethod = "POST"
request.setValue(apiKey, forHTTPHeaderField: "x-api-key")
request.setValue("2025-10-01", forHTTPHeaderField: "anthropic-version")
request.setValue("application/json", forHTTPHeaderField: "Content-Type")
request.httpBody = try? JSONSerialization.data(withJSONObject: payload)

let task = URLSession.shared.dataTask(with: request) { data, response, error in
    if let error = error {
        print("Error: \(error)")
        return
    }
    
    if let data = data, let responseString = String(data: data, encoding: .utf8) {
        print(responseString)
    }
}

task.resume()

using System;
using System.Net.Http;
using System.Text;
using System.Threading.Tasks;

class Program
{
    static async Task Main(string[] args)
    {
        var url = "https://api.apimart.ai/v1/messages";
        var apiKey = Environment.GetEnvironmentVariable("API_KEY");

        var payload = @"{
            ""model"": ""claude-sonnet-4-6"",
            ""max_tokens"": 1024,
            ""messages"": [
                {
                    ""role"": ""user"",
                    ""content"": ""Hello, world""
                }
            ]
        }";

        using var client = new HttpClient();
        client.DefaultRequestHeaders.Add("x-api-key", apiKey);
        client.DefaultRequestHeaders.Add("anthropic-version", "2025-10-01");

        var content = new StringContent(payload, Encoding.UTF8, "application/json");
        var response = await client.PostAsync(url, content);
        var result = await response.Content.ReadAsStringAsync();

        Console.WriteLine(result);
    }
}

#include <stdio.h>
#include <curl/curl.h>
#include <stdlib.h>

int main(void) {
    CURL *curl;
    CURLcode res;
    const char *api_key = getenv("API_KEY");

    curl_global_init(CURL_GLOBAL_DEFAULT);
    curl = curl_easy_init();

    if(curl) {
        const char *url = "https://api.apimart.ai/v1/messages";
        const char *payload = "{"
            "\"model\":\"claude-sonnet-4-6\","
            "\"max_tokens\":1024,"
            "\"messages\":[{\"role\":\"user\",\"content\":\"Hello, world\"}]"
        "}";

        char auth_header[256];
        snprintf(auth_header, sizeof(auth_header), "x-api-key: %s", api_key);

        struct curl_slist *headers = NULL;
        headers = curl_slist_append(headers, auth_header);
        headers = curl_slist_append(headers, "anthropic-version: 2025-10-01");
        headers = curl_slist_append(headers, "Content-Type: application/json");

        curl_easy_setopt(curl, CURLOPT_URL, url);
        curl_easy_setopt(curl, CURLOPT_POSTFIELDS, payload);
        curl_easy_setopt(curl, CURLOPT_HTTPHEADER, headers);

        res = curl_easy_perform(curl);

        if(res != CURLE_OK) {
            fprintf(stderr, "curl_easy_perform() failed: %s\n",
                    curl_easy_strerror(res));
        }

        curl_slist_free_all(headers);
        curl_easy_cleanup(curl);
    }

    curl_global_cleanup();
    return 0;
}

#import <Foundation/Foundation.h>

int main(int argc, const char * argv[]) {
    @autoreleasepool {
        NSURL *url = [NSURL URLWithString:@"https://api.apimart.ai/v1/messages"];
        NSString *apiKey = [NSProcessInfo processInfo].environment[@"API_KEY"];
        
        NSDictionary *payload = @{
            @"model": @"claude-sonnet-4-6",
            @"max_tokens": @1024,
            @"messages": @[
                @{
                    @"role": @"user",
                    @"content": @"Hello, world"
                }
            ]
        };
        
        NSError *error;
        NSData *jsonData = [NSJSONSerialization dataWithJSONObject:payload
                                                          options:0
                                                            error:&error];
        
        NSMutableURLRequest *request = [NSMutableURLRequest requestWithURL:url];
        [request setHTTPMethod:@"POST"];
        [request setValue:apiKey forHTTPHeaderField:@"x-api-key"];
        [request setValue:@"2025-10-01" forHTTPHeaderField:@"anthropic-version"];
        [request setValue:@"application/json" forHTTPHeaderField:@"Content-Type"];
        [request setHTTPBody:jsonData];
        
        NSURLSessionDataTask *task = [[NSURLSession sharedSession] 
            dataTaskWithRequest:request
            completionHandler:^(NSData *data, NSURLResponse *response, NSError *error) {
                if (error) {
                    NSLog(@"Error: %@", error);
                    return;
                }
                NSString *result = [[NSString alloc] initWithData:data 
                                                        encoding:NSUTF8StringEncoding];
                NSLog(@"%@", result);
            }];
        
        [task resume];
        [[NSRunLoop mainRunLoop] run];
    }
    return 0;
}

(* Requires cohttp and yojson libraries *)
open Lwt
open Cohttp
open Cohttp_lwt_unix

let url = "https://api.apimart.ai/v1/messages"
let api_key = Sys.getenv "API_KEY"

let payload = {|{
  "model": "claude-sonnet-4-6",
  "max_tokens": 1024,
  "messages": [
    {
      "role": "user",
      "content": "Hello, world"
    }
  ]
}|}

let () =
  let headers = Header.init ()
    |> fun h -> Header.add h "x-api-key" api_key
    |> fun h -> Header.add h "anthropic-version" "2025-10-01"
    |> fun h -> Header.add h "Content-Type" "application/json"
  in
  let body = Cohttp_lwt.Body.of_string payload in
  
  let response = Client.post ~headers ~body (Uri.of_string url) >>= fun (resp, body) ->
    body |> Cohttp_lwt.Body.to_string >|= fun body_str ->
    print_endline body_str
  in
  Lwt_main.run response

import 'dart:convert';
import 'dart:io';
import 'package:http/http.dart' as http;

void main() async {
  final url = Uri.parse('https://api.apimart.ai/v1/messages');
  final apiKey = Platform.environment['API_KEY'];
  
  final payload = {
    'model': 'claude-sonnet-4-6',
    'max_tokens': 1024,
    'messages': [
      {
        'role': 'user',
        'content': 'Hello, world'
      }
    ]
  };
  
  final response = await http.post(
    url,
    headers: {
      'x-api-key': apiKey!,
      'anthropic-version': '2025-10-01',
      'Content-Type': 'application/json',
    },
    body: jsonEncode(payload),
  );
  
  print(response.body);
}

library(httr)
library(jsonlite)

url <- "https://api.apimart.ai/v1/messages"
api_key <- Sys.getenv("API_KEY")

payload <- list(
  model = "claude-sonnet-4-6",
  max_tokens = 1024,
  messages = list(
    list(
      role = "user",
      content = "Hello, world"
    )
  )
)

response <- POST(
  url,
  add_headers(
    `x-api-key` = api_key,
    `anthropic-version` = "2025-10-01",
    `Content-Type` = "application/json"
  ),
  body = toJSON(payload, auto_unbox = TRUE),
  encode = "raw"
)

cat(content(response, "text"))

{
  "model": "claude-sonnet-4-6",
  "id": "msg_011CdfeHuC728oxqaLrRNbcB",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "type": "text",
      "text": "Hello! I'm Claude. Nice to meet you."
    }
  ],
  "stop_reason": "end_turn",
  "stop_sequence": null,
  "stop_details": null,
  "usage": {
    "input_tokens": 12,
    "cache_creation_input_tokens": 0,
    "cache_read_input_tokens": 0,
    "cache_creation": {
      "ephemeral_5m_input_tokens": 0,
      "ephemeral_1h_input_tokens": 0
    },
    "output_tokens": 18,
    "service_tier": "standard",
    "inference_geo": "global"
  }
}

{
  "error": {
    "code": "model_not_found",
    "message": "model not found (request id: 20260803181903172761862QzohKPXF)",
    "param": "",
    "type": "apimart_error"
  }
}

{
  "error": {
    "code": "",
    "message": "Invalid API key (request id: 20260803181903172761862QzohKPXF)",
    "param": "",
    "type": "apimart_error"
  }
}

{
  "error": {
    "code": "",
    "message": "Insufficient balance (request id: 20260803181903172761862QzohKPXF)",
    "param": "",
    "type": "apimart_error"
  }
}

{
  "error": {
    "code": "",
    "message": "Too many requests (request id: 20260803181903172761862QzohKPXF)",
    "param": "",
    "type": "apimart_error"
  }
}

{
  "error": {
    "code": "",
    "message": "Internal server error (request id: 20260803181903172761862QzohKPXF)",
    "param": "",
    "type": "apimart_error"
  }
}

Die beiden APIs nicht vermischen: /v1/* ist die Inference-API (dieses Dokument, Upstream 1:1, ohne Wrapper); /api/* ist die Management-API (Guthaben/Logs usw., Antwort {success, message, data}). Wenn an anderer Stelle steht, dass /v1/messages {code, data} zurückgibt, gilt dieses Dokument.

curl https://api.apimart.ai/v1/messages \
  -H "x-api-key: $API_KEY" \
  -H "anthropic-version: 2025-10-01" \
  -H "content-type: application/json" \
  -d '{
    "model": "claude-sonnet-4-6",
    "max_tokens": 1024,
    "messages": [
      {"role": "user", "content": "Hello, world"}
    ]
  }'

import anthropic

client = anthropic.Anthropic(
    api_key="YOUR_API_KEY",
    base_url="https://api.apimart.ai"
)

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

print(message.content)

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

const client = new Anthropic({
  apiKey: process.env.API_KEY,
  baseURL: 'https://api.apimart.ai'
});

const message = await client.messages.create({
  model: 'claude-sonnet-4-6',
  max_tokens: 1024,
  messages: [
    { role: 'user', content: 'Hello, world' }
  ]
});

console.log(message.content);

package main

import (
    "bytes"
    "encoding/json"
    "fmt"
    "io/ioutil"
    "net/http"
    "os"
)

func main() {
    url := "https://api.apimart.ai/v1/messages"

    payload := map[string]interface{}{
        "model": "claude-sonnet-4-6",
        "max_tokens": 1024,
        "messages": []map[string]string{
            {
                "role":    "user",
                "content": "Hello, world",
            },
        },
    }

    jsonData, _ := json.Marshal(payload)

    req, _ := http.NewRequest("POST", url, bytes.NewBuffer(jsonData))
    req.Header.Set("x-api-key", os.Getenv("API_KEY"))
    req.Header.Set("anthropic-version", "2025-10-01")
    req.Header.Set("Content-Type", "application/json")

    client := &http.Client{}
    resp, err := client.Do(req)
    if err != nil {
        panic(err)
    }
    defer resp.Body.Close()

    body, _ := ioutil.ReadAll(resp.Body)
    fmt.Println(string(body))
}

import java.net.http.HttpClient;
import java.net.http.HttpRequest;
import java.net.http.HttpResponse;
import java.net.URI;

public class Main {
    public static void main(String[] args) throws Exception {
        String url = "https://api.apimart.ai/v1/messages";
        String apiKey = System.getenv("API_KEY");

        String payload = """
        {
          "model": "claude-sonnet-4-6",
          "max_tokens": 1024,
          "messages": [
            {
              "role": "user",
              "content": "Hello, world"
            }
          ]
        }
        """;

        HttpClient client = HttpClient.newHttpClient();
        HttpRequest request = HttpRequest.newBuilder()
            .uri(URI.create(url))
            .header("x-api-key", apiKey)
            .header("anthropic-version", "2025-10-01")
            .header("Content-Type", "application/json")
            .POST(HttpRequest.BodyPublishers.ofString(payload))
            .build();

        HttpResponse<String> response = client.send(request,
            HttpResponse.BodyHandlers.ofString());

        System.out.println(response.body());
    }
}

<?php

$url = "https://api.apimart.ai/v1/messages";
$apiKey = getenv('API_KEY');

$payload = [
    "model" => "claude-sonnet-4-6",
    "max_tokens" => 1024,
    "messages" => [
        [
            "role" => "user",
            "content" => "Hello, world"
        ]
    ]
];

$ch = curl_init($url);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
curl_setopt($ch, CURLOPT_POST, true);
curl_setopt($ch, CURLOPT_POSTFIELDS, json_encode($payload));
curl_setopt($ch, CURLOPT_HTTPHEADER, [
    "x-api-key: " . $apiKey,
    "anthropic-version: 2025-10-01",
    "Content-Type: application/json"
]);

$response = curl_exec($ch);
curl_close($ch);

echo $response;
?>

require 'net/http'
require 'json'
require 'uri'

url = URI("https://api.apimart.ai/v1/messages")
api_key = ENV['API_KEY']

payload = {
  model: "claude-sonnet-4-6",
  max_tokens: 1024,
  messages: [
    {
      role: "user",
      content: "Hello, world"
    }
  ]
}

http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true

request = Net::HTTP::Post.new(url)
request["x-api-key"] = api_key
request["anthropic-version"] = "2025-10-01"
request["Content-Type"] = "application/json"
request.body = payload.to_json

response = http.request(request)
puts response.body

import Foundation

let url = URL(string: "https://api.apimart.ai/v1/messages")!
let apiKey = ProcessInfo.processInfo.environment["API_KEY"] ?? ""

let payload: [String: Any] = [
    "model": "claude-sonnet-4-6",
    "max_tokens": 1024,
    "messages": [
        [
            "role": "user",
            "content": "Hello, world"
        ]
    ]
]

var request = URLRequest(url: url)
request.httpMethod = "POST"
request.setValue(apiKey, forHTTPHeaderField: "x-api-key")
request.setValue("2025-10-01", forHTTPHeaderField: "anthropic-version")
request.setValue("application/json", forHTTPHeaderField: "Content-Type")
request.httpBody = try? JSONSerialization.data(withJSONObject: payload)

let task = URLSession.shared.dataTask(with: request) { data, response, error in
    if let error = error {
        print("Error: \(error)")
        return
    }
    
    if let data = data, let responseString = String(data: data, encoding: .utf8) {
        print(responseString)
    }
}

task.resume()

using System;
using System.Net.Http;
using System.Text;
using System.Threading.Tasks;

class Program
{
    static async Task Main(string[] args)
    {
        var url = "https://api.apimart.ai/v1/messages";
        var apiKey = Environment.GetEnvironmentVariable("API_KEY");

        var payload = @"{
            ""model"": ""claude-sonnet-4-6"",
            ""max_tokens"": 1024,
            ""messages"": [
                {
                    ""role"": ""user"",
                    ""content"": ""Hello, world""
                }
            ]
        }";

        using var client = new HttpClient();
        client.DefaultRequestHeaders.Add("x-api-key", apiKey);
        client.DefaultRequestHeaders.Add("anthropic-version", "2025-10-01");

        var content = new StringContent(payload, Encoding.UTF8, "application/json");
        var response = await client.PostAsync(url, content);
        var result = await response.Content.ReadAsStringAsync();

        Console.WriteLine(result);
    }
}

#include <stdio.h>
#include <curl/curl.h>
#include <stdlib.h>

int main(void) {
    CURL *curl;
    CURLcode res;
    const char *api_key = getenv("API_KEY");

    curl_global_init(CURL_GLOBAL_DEFAULT);
    curl = curl_easy_init();

    if(curl) {
        const char *url = "https://api.apimart.ai/v1/messages";
        const char *payload = "{"
            "\"model\":\"claude-sonnet-4-6\","
            "\"max_tokens\":1024,"
            "\"messages\":[{\"role\":\"user\",\"content\":\"Hello, world\"}]"
        "}";

        char auth_header[256];
        snprintf(auth_header, sizeof(auth_header), "x-api-key: %s", api_key);

        struct curl_slist *headers = NULL;
        headers = curl_slist_append(headers, auth_header);
        headers = curl_slist_append(headers, "anthropic-version: 2025-10-01");
        headers = curl_slist_append(headers, "Content-Type: application/json");

        curl_easy_setopt(curl, CURLOPT_URL, url);
        curl_easy_setopt(curl, CURLOPT_POSTFIELDS, payload);
        curl_easy_setopt(curl, CURLOPT_HTTPHEADER, headers);

        res = curl_easy_perform(curl);

        if(res != CURLE_OK) {
            fprintf(stderr, "curl_easy_perform() failed: %s\n",
                    curl_easy_strerror(res));
        }

        curl_slist_free_all(headers);
        curl_easy_cleanup(curl);
    }

    curl_global_cleanup();
    return 0;
}

#import <Foundation/Foundation.h>

int main(int argc, const char * argv[]) {
    @autoreleasepool {
        NSURL *url = [NSURL URLWithString:@"https://api.apimart.ai/v1/messages"];
        NSString *apiKey = [NSProcessInfo processInfo].environment[@"API_KEY"];
        
        NSDictionary *payload = @{
            @"model": @"claude-sonnet-4-6",
            @"max_tokens": @1024,
            @"messages": @[
                @{
                    @"role": @"user",
                    @"content": @"Hello, world"
                }
            ]
        };
        
        NSError *error;
        NSData *jsonData = [NSJSONSerialization dataWithJSONObject:payload
                                                          options:0
                                                            error:&error];
        
        NSMutableURLRequest *request = [NSMutableURLRequest requestWithURL:url];
        [request setHTTPMethod:@"POST"];
        [request setValue:apiKey forHTTPHeaderField:@"x-api-key"];
        [request setValue:@"2025-10-01" forHTTPHeaderField:@"anthropic-version"];
        [request setValue:@"application/json" forHTTPHeaderField:@"Content-Type"];
        [request setHTTPBody:jsonData];
        
        NSURLSessionDataTask *task = [[NSURLSession sharedSession] 
            dataTaskWithRequest:request
            completionHandler:^(NSData *data, NSURLResponse *response, NSError *error) {
                if (error) {
                    NSLog(@"Error: %@", error);
                    return;
                }
                NSString *result = [[NSString alloc] initWithData:data 
                                                        encoding:NSUTF8StringEncoding];
                NSLog(@"%@", result);
            }];
        
        [task resume];
        [[NSRunLoop mainRunLoop] run];
    }
    return 0;
}

(* Requires cohttp and yojson libraries *)
open Lwt
open Cohttp
open Cohttp_lwt_unix

let url = "https://api.apimart.ai/v1/messages"
let api_key = Sys.getenv "API_KEY"

let payload = {|{
  "model": "claude-sonnet-4-6",
  "max_tokens": 1024,
  "messages": [
    {
      "role": "user",
      "content": "Hello, world"
    }
  ]
}|}

let () =
  let headers = Header.init ()
    |> fun h -> Header.add h "x-api-key" api_key
    |> fun h -> Header.add h "anthropic-version" "2025-10-01"
    |> fun h -> Header.add h "Content-Type" "application/json"
  in
  let body = Cohttp_lwt.Body.of_string payload in
  
  let response = Client.post ~headers ~body (Uri.of_string url) >>= fun (resp, body) ->
    body |> Cohttp_lwt.Body.to_string >|= fun body_str ->
    print_endline body_str
  in
  Lwt_main.run response

import 'dart:convert';
import 'dart:io';
import 'package:http/http.dart' as http;

void main() async {
  final url = Uri.parse('https://api.apimart.ai/v1/messages');
  final apiKey = Platform.environment['API_KEY'];
  
  final payload = {
    'model': 'claude-sonnet-4-6',
    'max_tokens': 1024,
    'messages': [
      {
        'role': 'user',
        'content': 'Hello, world'
      }
    ]
  };
  
  final response = await http.post(
    url,
    headers: {
      'x-api-key': apiKey!,
      'anthropic-version': '2025-10-01',
      'Content-Type': 'application/json',
    },
    body: jsonEncode(payload),
  );
  
  print(response.body);
}

library(httr)
library(jsonlite)

url <- "https://api.apimart.ai/v1/messages"
api_key <- Sys.getenv("API_KEY")

payload <- list(
  model = "claude-sonnet-4-6",
  max_tokens = 1024,
  messages = list(
    list(
      role = "user",
      content = "Hello, world"
    )
  )
)

response <- POST(
  url,
  add_headers(
    `x-api-key` = api_key,
    `anthropic-version` = "2025-10-01",
    `Content-Type` = "application/json"
  ),
  body = toJSON(payload, auto_unbox = TRUE),
  encode = "raw"
)

cat(content(response, "text"))

{
  "model": "claude-sonnet-4-6",
  "id": "msg_011CdfeHuC728oxqaLrRNbcB",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "type": "text",
      "text": "Hello! I'm Claude. Nice to meet you."
    }
  ],
  "stop_reason": "end_turn",
  "stop_sequence": null,
  "stop_details": null,
  "usage": {
    "input_tokens": 12,
    "cache_creation_input_tokens": 0,
    "cache_read_input_tokens": 0,
    "cache_creation": {
      "ephemeral_5m_input_tokens": 0,
      "ephemeral_1h_input_tokens": 0
    },
    "output_tokens": 18,
    "service_tier": "standard",
    "inference_geo": "global"
  }
}

{
  "error": {
    "code": "model_not_found",
    "message": "model not found (request id: 20260803181903172761862QzohKPXF)",
    "param": "",
    "type": "apimart_error"
  }
}

{
  "error": {
    "code": "",
    "message": "Invalid API key (request id: 20260803181903172761862QzohKPXF)",
    "param": "",
    "type": "apimart_error"
  }
}

{
  "error": {
    "code": "",
    "message": "Insufficient balance (request id: 20260803181903172761862QzohKPXF)",
    "param": "",
    "type": "apimart_error"
  }
}

{
  "error": {
    "code": "",
    "message": "Too many requests (request id: 20260803181903172761862QzohKPXF)",
    "param": "",
    "type": "apimart_error"
  }
}

{
  "error": {
    "code": "",
    "message": "Internal server error (request id: 20260803181903172761862QzohKPXF)",
    "param": "",
    "type": "apimart_error"
  }
}

Autorisierung

Zwei Authentifizierungsmethoden werden unterstützt — wählen Sie eine:

string

Authentifizierungs-Header im Anthropic-StilBesuchen Sie die Seite zur API-Key-Verwaltung, um Ihren API-Key zu erhalten

x-api-key: YOUR_API_KEY

string

Bearer-Token-Authentifizierung (Alternative zu x-api-key)

Authorization: Bearer YOUR_API_KEY

string

API-Version (optional; funktioniert auch ohne Header)Empfohlen, um später leichter auf den offiziellen Anthropic-Endpunkt zu migrieren:Beispiel: 2025-10-01

Body

string

Standard:"claude-sonnet-4-6"

erforderlich

Model name

claude-opus-4-8 - Claude Opus 4.8 flagship model
claude-opus-4-7 - Claude Opus 4.7 flagship model
claude-opus-4-6 - Claude Opus 4.6 flagship model
claude-sonnet-4-6 - Claude Sonnet 4.6 balanced version
claude-opus-4-5-20251101 - Claude Opus 4.5 model

array

erforderlich

Liste von NachrichtenArray von Nachrichten, auf deren Grundlage das Modell die nächste Antwort generiert. Jede Nachricht enthält die Felder role und content.💡 Schnellausfüllen (Try-it-Bereich):

Klicken Sie auf „+ Add an item”, um eine Nachricht hinzuzufügen
Eingabe role: user (Benutzernachricht) oder assistant (KI-Antwort, für Mehrfach-Dialoge)
Eingabe content: der Text Ihrer Nachricht

Anzeigen Felddetails

string

Standard:"user"

erforderlich

RollentypOptionen: user (Benutzernachricht), assistant (KI-Antwort, für Mehrfach-Dialoge und Prefilling)Hinweis: Die Claude API verwendet für Systemanweisungen einen separaten Parameter system, nicht in messages

string

erforderlich

NachrichteninhaltTextinhalt der Nachricht

Einzelne Benutzernachricht:

[{"role": "user", "content": "Hello, Claude"}]

Mehrfach-Dialog:

[
  {"role": "user", "content": "Hello there."},
  {"role": "assistant", "content": "Hi, I'm Claude. How can I help you?"},
  {"role": "user", "content": "Can you explain LLMs in plain English?"}
]

Vorbefüllte Assistant-Antwort:

[
  {"role": "user", "content": "What's the Greek name for Sun? (A) Sol (B) Helios (C) Sun"},
  {"role": "assistant", "content": "The best answer is ("}
]

integer

erforderlich

Maximale Anzahl zu generierender Tokens (pflicht, wie bei Anthropic)Maximale Anzahl an Tokens, bevor die Generierung stoppt. Das Modell kann bereits vor Erreichen dieser Grenze stoppen.Verschiedene Modelle haben unterschiedliche Maximalwerte. Minimum: 1

object

Extended-thinking-KonfigurationBei Aktivierung kann die Antwort-content thinking-Blöcke enthalten. Empfohlen ist der Standard-Modellname + dieser Parameter statt plattformseitiger -thinking-Aliase — erleichtert die Migration zum offiziellen Endpunkt ohne Codeänderungen.Wenn Thinking-Blöcke in Mehrfachdialogen zurückgesendet werden, muss signature unverändert mitgeschickt werden, sonst lehnt der Upstream ab.

string | array

SystemanweisungSystemanweisungen legen Claudes Rolle, Persönlichkeit, Ziele und Anweisungen fest.String-Format:

{
  "system": "You are a professional Python programming tutor"
}

Strukturiertes Format:

{
  "system": [
    {
      "type": "text",
      "text": "You are a professional Python programming tutor"
    }
  ]
}

number

Temperatur-Parameter, Bereich 0–1Steuert die Zufälligkeit der Ausgabe:

Niedrige Werte (z. B. 0.2): deterministischer, konservativer
Hohe Werte (z. B. 0.8): zufälliger, kreativer

Standard: 1.0

number

Nucleus-Sampling-Parameter, Bereich 0–1Verwendet Nucleus-Sampling. Es wird empfohlen, entweder temperature oder top_p zu verwenden, nicht beides.Standard: 1.0

integer

Top-K-SamplingSampling nur aus den Top-K-Optionen, entfernt „long tail”-Antworten mit niedriger Wahrscheinlichkeit.Nur für fortgeschrittene Anwendungsfälle empfohlen.

boolean

Streaming aktivierenBei true werden Server-Sent Events (SSE) zur Streaming-Übertragung der Antworten verwendet.Standard: false

array

Stopp-SequenzenBenutzerdefinierte Textsequenzen, bei denen das Modell die Generierung stoppt.Maximal 4 Sequenzen.Beispiel: ["\n\nHuman:", "\n\nAssistant:"]

object

MetadatenMetadaten-Objekt für die Anfrage.Enthält:

user_id: Benutzer-Identifikator

array

Tool-DefinitionenListe der Tools, die das Modell zur Erledigung von Aufgaben verwenden kann.Beispiel für ein Funktions-Tool:

{
  "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"
          },
          "unit": {
            "type": "string",
            "enum": ["celsius", "fahrenheit"],
            "description": "Temperature unit"
          }
        },
        "required": ["location"]
      }
    }
  ]
}

Unterstützte Tool-Typen:

Benutzerdefinierte Funktions-Tools
Computer-Use-Tool (computer_20241022)
Texteditor-Tool (text_editor_20241022)
Bash-Tool (bash_20241022)

object

Tool-AuswahlstrategieSteuert, wie das Modell Tools verwendet:

{"type": "auto"}: automatische Entscheidung (Standard)
{"type": "any"}: muss ein Tool verwenden
{"type": "tool", "name": "tool_name"}: spezifisches Tool verwenden

Response

string

Eindeutiger Nachrichten-IdentifikatorBeispiel: "msg_013Zva2CMHLNnXjNJJKqJ2EF"

string

ObjekttypImmer "message"

string

RolleImmer "assistant"

array

Array von Content-BlöckenIn content wird der Blocktyp über type unterschieden. Eine Antwort kann mehrere Blöcke enthalten (z. B. bei aktiviertem Thinking: thinking + text).text-Block:

{ "type": "text", "text": "OK" }

tool_use-Block:

{
  "type": "tool_use",
  "id": "toolu_01QgsazxKXSfQVj9Q1XxjYXo",
  "name": "get_weather",
  "input": { "city": "Beijing" },
  "caller": { "type": "direct" }
}

caller ist ein neues Upstream-Feld und noch nicht in der offiziellen Dokumentation enthalten. Beim Parsen ignorieren.

thinking-Block (erscheint, wenn der Request-Body den Parameter thinking enthält):

{
  "type": "thinking",
  "thinking": "推理过程文本...",
  "signature": "<约 500+ 字符的签名串>"
}

Wenn Thinking-Blöcke in Mehrfachdialogen zurückgesendet werden, muss signature unverändert mitgeschickt werden, sonst lehnt der Upstream ab.

Nicht annehmen, dass content[0] Text ist. Bei aktiviertem Thinking kann content[0] ein Thinking-Block sein. Durchlaufen und filtern:

text = "".join(b.text for b in resp.content if b.type == "text")

string

Das Modell, das die Anfrage bearbeitet hatBeispiel: "claude-sonnet-4-6"

string

Stopp-GrundMögliche Werte:

end_turn: natürliche Beendigung
max_tokens: maximale Token-Anzahl erreicht
stop_sequence: Stopp-Sequenz erreicht
tool_use: Tool aufgerufen

string | null

Ausgelöste Stopp-SequenzBei Stopp durch eine Stopp-Sequenz deren Inhalt; sonst null

object | null

Neueres Anthropic-Feld; bei normalen Anfragen null

object

Statistik zur Token-Nutzung (vollständige Struktur bei Nicht-Streaming)

Anzeigen Eigenschaften

integer

Anzahl der Eingabe-Tokens

integer

Anzahl der Ausgabe-Tokens (enthält bereits Thinking-Tokens; bei Abrechnung nicht doppelt zählen)

integer

Cache-Schreib-Tokens

integer

Cache-Treffer-Tokens

object

{ ephemeral_5m_input_tokens, ephemeral_1h_input_tokens }

string

z. B. "standard"

string

z. B. "global"

object

Kann bei Thinking erscheinen: { thinking_tokens: int }

Anwendungsbeispiele

Einfacher Dialog

import anthropic

client = anthropic.Anthropic(
    api_key="YOUR_API_KEY",
    base_url="https://api.apimart.ai"
)

message = client.messages.create(
    model="claude-sonnet-4-6",
    max_tokens=1024,
    messages=[
        {"role": "user", "content": "Explain quantum computing basics"}
    ]
)

print(message.content[0].text)

Mehrfach-Dialog

messages = [
    {"role": "user", "content": "What is machine learning?"},
    {"role": "assistant", "content": "Machine learning is a branch of AI..."},
    {"role": "user", "content": "Can you give a practical example?"}
]

message = client.messages.create(
    model="claude-sonnet-4-6",
    max_tokens=1024,
    messages=messages
)

Verwendung von Systemanweisungen

message = client.messages.create(
    model="claude-sonnet-4-6",
    max_tokens=1024,
    system="You are a senior Python developer expert in code review and optimization.",
    messages=[
        {"role": "user", "content": "How to optimize this code?\n\n[code]"}
    ]
)

Streaming-Antwort

with client.messages.stream(
    model="claude-sonnet-4-6",
    max_tokens=1024,
    messages=[
        {"role": "user", "content": "Write a short essay about AI"}
    ]
) as stream:
    for text in stream.text_stream:
        print(text, end="", flush=True)

Tool-Nutzung

tools = [
    {
        "name": "get_stock_price",
        "description": "Get real-time stock price",
        "input_schema": {
            "type": "object",
            "properties": {
                "ticker": {
                    "type": "string",
                    "description": "Stock ticker symbol, e.g., AAPL"
                }
            },
            "required": ["ticker"]
        }
    }
]

message = client.messages.create(
    model="claude-sonnet-4-6",
    max_tokens=1024,
    tools=tools,
    messages=[
        {"role": "user", "content": "What's Tesla's stock price?"}
    ]
)

# Handle tool calls
if message.stop_reason == "tool_use":
    tool_use = next(block for block in message.content if block.type == "tool_use")
    print(f"Calling tool: {tool_use.name}")
    print(f"Arguments: {tool_use.input}")

Bildverständnis

message = client.messages.create(
    model="claude-sonnet-4-6",
    max_tokens=1024,
    messages=[
        {
            "role": "user",
            "content": [
                {
                    "type": "image",
                    "source": {
                        "type": "url",
                        "url": "https://example.com/image.jpg"
                    }
                },
                {
                    "type": "text",
                    "text": "Describe this image"
                }
            ]
        }
    ]
)

Bild im Base64-Format

import base64

with open("image.jpg", "rb") as image_file:
    image_data = base64.b64encode(image_file.read()).decode("utf-8")

message = client.messages.create(
    model="claude-sonnet-4-6",
    max_tokens=1024,
    messages=[
        {
            "role": "user",
            "content": [
                {
                    "type": "image",
                    "source": {
                        "type": "base64",
                        "media_type": "image/jpeg",
                        "data": image_data
                    }
                },
                {
                    "type": "text",
                    "text": "Analyze this image"
                }
            ]
        }
    ]
)

Best Practices

1. Prompt Engineering

Klare Rollendefinition:

system = """You are an experienced data scientist specializing in:
- Statistical analysis and data visualization
- Machine learning model development
- Python and R programming
Provide professional, accurate advice."""

Strukturierte Ausgabe:

message = "Please return the analysis results in JSON format with summary, key_findings, and recommendations fields."

2. Fehlerbehandlung

from anthropic import APIError, RateLimitError

try:
    message = client.messages.create(
        model="claude-sonnet-4-6",
        max_tokens=1024,
        messages=[{"role": "user", "content": "Hello"}]
    )
except RateLimitError:
    print("Rate limit exceeded, please retry later")
except APIError as e:
    print(f"API error: {e}")

3. Token-Optimierung

# Use shorter prompts
messages = [
    {"role": "user", "content": "Summarize key points:\n\n[long text]"}
]

# Limit output length
message = client.messages.create(
    model="claude-sonnet-4-6",
    max_tokens=500,  # Limit output
    messages=messages
)

4. Vorbefüllung von Antworten

# Guide model to specific format
messages = [
    {"role": "user", "content": "List 5 Python best practices"},
    {"role": "assistant", "content": "Here are 5 Python best practices:\n\n1."}
]

message = client.messages.create(
    model="claude-sonnet-4-6",
    max_tokens=1024,
    messages=messages
)

Verarbeitung von Streaming-Antworten

Streaming in Python

import anthropic

client = anthropic.Anthropic(
    api_key="YOUR_API_KEY",
    base_url="https://api.apimart.ai"
)

with client.messages.stream(
    model="claude-sonnet-4-6",
    max_tokens=1024,
    messages=[
        {"role": "user", "content": "Write a Python decorator example"}
    ]
) as stream:
    for text in stream.text_stream:
        print(text, end="", flush=True)

Streaming in JavaScript

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

const client = new Anthropic({
  apiKey: process.env.API_KEY,
  baseURL: 'https://api.apimart.ai'
});

const stream = await client.messages.stream({
  model: 'claude-sonnet-4-6',
  max_tokens: 1024,
  messages: [
    { role: 'user', content: 'Write a React component example' }
  ]
});

for await (const chunk of stream) {
  if (chunk.type === 'content_block_delta' && 
      chunk.delta.type === 'text_delta') {
    process.stdout.write(chunk.delta.text);
  }
}

Plattformunterschiede und Integrationshinweise

Antwort ohne Wrapper

Bei Erfolg gibt POST /v1/messages das Anthropic-Message-Objekt direkt zurück — ohne äußere {code, data}-Hülle. Damit sind offizielles SDK, Claude Code, Cline usw. 1:1 kompatibel.

Fehlerformat (einziger wesentlicher Unterschied zum Offiziellen)

{
  "error": {
    "code": "model_not_found",
    "message": "... (request id: ...)",
    "param": "",
    "type": "apimart_error"
  }
}

Im Vergleich zu Anthropic: Top-Level fehlt "type": "error"; error.type ist fest apimart_error, nicht semantische Typen wie invalid_request_error. Integrationsempfehlung: Retry-Logik nicht an error.type koppeln, sondern HTTP-Status + error.code nutzen:

Status	Bedeutung	Empfohlene Aktion
400	Parameterfehler	Nicht erneut versuchen, Request-Body prüfen
401	Ungültiger Key	Nicht erneut versuchen
402	Guthaben unzureichend	Nicht erneut versuchen, Aufladen vorschlagen
429	Rate-Limit	Nach Backoff erneut versuchen
5xx	Upstream-/Gateway-Fehler	Mit exponentiellem Backoff erneut versuchen

Bei Support-Anfragen: Request-ID am Ende von error.message und Response-Header x-oneapi-request-id angeben.

Streaming-SSE

Request mit "stream": true. Ereignisreihenfolge wie offiziell: message_start → content_block_start → ping → content_block_delta (mehrfach) → content_block_stop → message_delta → message_stop ⚠️ usage-Struktur unterscheidet sich zwischen Stream und Non-Stream: message_delta.usage hat typischerweise nur 4 Token-Felder, ohne cache_creation, service_tier, inference_geo. Getrennt parsen oder alle Felder optional behandeln.

Nicht implementierte Endpunkte

POST /v1/messages/count_tokens ist nicht implementiert und liefert 404. client.messages.count_tokens() des offiziellen SDK schlägt fehl. Token-Schätzung lokal vornehmen oder usage.input_tokens aus der Antwort lesen.

Unbekannte Felder ignorieren

Diese API reicht Upstream durch; Anthropic kann jederzeit Felder ergänzen (z. B. stop_details, inference_geo, caller, output_tokens_details). Keine strenge Schema-Validierung:

Go: kein DisallowUnknownFields()
Pydantic: kein extra="forbid"
TypeScript / Zod: .passthrough() statt .strict()

Empfehlung zu Modellnamen

Gleichnamige Modelle mit Suffix -thinking sind Plattform-Aliase. Empfohlen ist der Standardname ohne Suffix + Parameter thinking im Request-Body — erleichtert die Migration zum offiziellen Endpunkt. Weitere Request-Body-Felder entsprechen dem Offiziellen: model, messages, max_tokens (pflicht), system, temperature, top_p, top_k, stop_sequences, stream, tools, tool_choice, thinking, metadata. Semantik gemäß Anthropic Messages API.

Wichtige Hinweise

API-Key-Sicherheit:
- Speichern Sie API-Keys in Umgebungsvariablen
- Schreiben Sie Keys niemals fest in den Quellcode
- Rotieren Sie Keys regelmäßig
Rate-Limiting:
- Beachten Sie die API-Rate-Limits
- Implementieren Sie Retry-Mechanismen (nach HTTP-Status)
- Verwenden Sie exponentielles Backoff
Token-Verwaltung:
- Überwachen Sie die Token-Nutzung (usage lesen)
- Optimieren Sie die Prompt-Länge
- Verwenden Sie passende max_tokens-Werte
- Bei Thinking enthält output_tokens Thinking bereits — nicht doppelt abrechnen
Modellauswahl:
- Opus: komplexe Aufgaben, die tiefes Denken erfordern
- Sonnet: ausgewogene Leistung und Kosten
- Haiku: schnelle Antwort, einfache Aufgaben
Content-Parsing:
- content durchlaufen und type == "text" wählen; nicht hart content[0].text annehmen
- Liefert das Modell JSON in Markdown-Codeblöcken, ist das Modellausgabe, kein API-Wrapper (siehe FAQ unten)
Inhaltsfilterung:
- Validieren Sie Benutzereingaben
- Filtern Sie sensible Informationen
- Implementieren Sie Inhaltsmoderation

FAQ

Im Antwort-`content` ist text ein ```json ... ```-Codeblock — wie entfernen?

Das ist kein API-Strukturproblem. Im Feld text steht der rohe Modelloutput: Hält das Modell JSON für gewünscht, packt es ihn in einen Markdown-Codeblock. Die API schreibt den Modelloutput nicht um und soll das auch nicht. Für saubere strukturierte Daten gibt es drei richtige Ansätze (empfohlen absteigend):

Structured Output per tools erzwingen — am zuverlässigsten; input ist bereits das geparste Objekt:

{
  "tools": [{
    "name": "emit_result",
    "input_schema": {
      "type": "object",
      "properties": { "answer": { "type": "string" } }
    }
  }],
  "tool_choice": { "type": "tool", "name": "emit_result" }
}

Assistant-Nachricht prefillen, damit das Modell bei { weiterschreibt:

{
  "messages": [
    { "role": "user", "content": "..." },
    { "role": "assistant", "content": "{" }
  ]
}

Im System-Prompt klar fordern: „Nur JSON ausgeben, keinen Markdown-Codeblock“.

Code-Fences per Regex abzustreifen wird nicht empfohlen — fehlt die Fence gelegentlich, schlägt das Parsen fehl.

Übersicht

Modelle

Textserie

Bildserie

Videoserie

Audioserie

Moderationsserie

Upload-Verwaltung

Aufgabenverwaltung

Kontoverwaltung

Claude Messages API

Autorisierung

Body

Response

Anwendungsbeispiele

Einfacher Dialog

Mehrfach-Dialog

Verwendung von Systemanweisungen

Streaming-Antwort

Tool-Nutzung

Bildverständnis

Bild im Base64-Format

Best Practices

1. Prompt Engineering

2. Fehlerbehandlung

3. Token-Optimierung

4. Vorbefüllung von Antworten

Verarbeitung von Streaming-Antworten

Streaming in Python

Streaming in JavaScript

Plattformunterschiede und Integrationshinweise

Antwort ohne Wrapper

Fehlerformat (einziger wesentlicher Unterschied zum Offiziellen)

Streaming-SSE

Nicht implementierte Endpunkte

Unbekannte Felder ignorieren

Empfehlung zu Modellnamen

Wichtige Hinweise

FAQ

Im Antwort-`content` ist text ein ```json ... ```-Codeblock — wie entfernen?

​Autorisierung

​Body

​Response

​Anwendungsbeispiele

​Einfacher Dialog

​Mehrfach-Dialog

​Verwendung von Systemanweisungen

​Streaming-Antwort

​Tool-Nutzung

​Bildverständnis

​Bild im Base64-Format

​Best Practices

​1. Prompt Engineering

​2. Fehlerbehandlung

​3. Token-Optimierung

​4. Vorbefüllung von Antworten

​Verarbeitung von Streaming-Antworten

​Streaming in Python

​Streaming in JavaScript

​Plattformunterschiede und Integrationshinweise

​Antwort ohne Wrapper

​Fehlerformat (einziger wesentlicher Unterschied zum Offiziellen)

​Streaming-SSE

​Nicht implementierte Endpunkte

​Unbekannte Felder ignorieren

​Empfehlung zu Modellnamen

​Wichtige Hinweise

​FAQ

​Im Antwort-content ist text ein ```json ... ```-Codeblock — wie entfernen?

Autorisierung

Body

Response

Anwendungsbeispiele

Einfacher Dialog

Mehrfach-Dialog

Verwendung von Systemanweisungen

Streaming-Antwort

Tool-Nutzung

Bildverständnis

Bild im Base64-Format

Best Practices

1. Prompt Engineering

2. Fehlerbehandlung

3. Token-Optimierung

4. Vorbefüllung von Antworten

Verarbeitung von Streaming-Antworten

Streaming in Python

Streaming in JavaScript

Plattformunterschiede und Integrationshinweise

Antwort ohne Wrapper

Fehlerformat (einziger wesentlicher Unterschied zum Offiziellen)

Streaming-SSE

Nicht implementierte Endpunkte

Unbekannte Felder ignorieren

Empfehlung zu Modellnamen

Wichtige Hinweise

FAQ

Im Antwort-`content` ist text ein ```json ... ```-Codeblock — wie entfernen?