Grok Chat-Vervollständigung (alle Modelle, vollständige Parameter)

curl --request POST \
  --url https://direct.evolink.ai/v1/chat/completions \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "grok-4.5",
  "messages": [
    {
      "role": "system",
      "content": "You are a concise assistant."
    },
    {
      "role": "user",
      "content": "Explain prompt caching in one sentence."
    }
  ],
  "stream": false,
  "max_tokens": 1024,
  "temperature": 0.7,
  "top_p": 0.95,
  "tools": [
    {
      "type": "function",
      "function": {}
    }
  ]
}
'

import requests

url = "https://direct.evolink.ai/v1/chat/completions"

payload = {
    "model": "grok-4.5",
    "messages": [
        {
            "role": "system",
            "content": "You are a concise assistant."
        },
        {
            "role": "user",
            "content": "Explain prompt caching in one sentence."
        }
    ],
    "stream": False,
    "max_tokens": 1024,
    "temperature": 0.7,
    "top_p": 0.95,
    "tools": [
        {
            "type": "function",
            "function": {}
        }
    ]
}
headers = {
    "Authorization": "Bearer <token>",
    "Content-Type": "application/json"
}

response = requests.post(url, json=payload, headers=headers)

print(response.text)

const options = {
  method: 'POST',
  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
  body: JSON.stringify({
    model: 'grok-4.5',
    messages: [
      {role: 'system', content: 'You are a concise assistant.'},
      {role: 'user', content: 'Explain prompt caching in one sentence.'}
    ],
    stream: false,
    max_tokens: 1024,
    temperature: 0.7,
    top_p: 0.95,
    tools: [{type: 'function', function: {}}]
  })
};

fetch('https://direct.evolink.ai/v1/chat/completions', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

<?php

$curl = curl_init();

curl_setopt_array($curl, [
  CURLOPT_URL => "https://direct.evolink.ai/v1/chat/completions",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 10,
  CURLOPT_TIMEOUT => 30,
  CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
  CURLOPT_CUSTOMREQUEST => "POST",
  CURLOPT_POSTFIELDS => json_encode([
    'model' => 'grok-4.5',
    'messages' => [
        [
                'role' => 'system',
                'content' => 'You are a concise assistant.'
        ],
        [
                'role' => 'user',
                'content' => 'Explain prompt caching in one sentence.'
        ]
    ],
    'stream' => false,
    'max_tokens' => 1024,
    'temperature' => 0.7,
    'top_p' => 0.95,
    'tools' => [
        [
                'type' => 'function',
                'function' => [
                                
                ]
        ]
    ]
  ]),
  CURLOPT_HTTPHEADER => [
    "Authorization: Bearer <token>",
    "Content-Type: application/json"
  ],
]);

$response = curl_exec($curl);
$err = curl_error($curl);

curl_close($curl);

if ($err) {
  echo "cURL Error #:" . $err;
} else {
  echo $response;
}

package main

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

func main() {

	url := "https://direct.evolink.ai/v1/chat/completions"

	payload := strings.NewReader("{\n  \"model\": \"grok-4.5\",\n  \"messages\": [\n    {\n      \"role\": \"system\",\n      \"content\": \"You are a concise assistant.\"\n    },\n    {\n      \"role\": \"user\",\n      \"content\": \"Explain prompt caching in one sentence.\"\n    }\n  ],\n  \"stream\": false,\n  \"max_tokens\": 1024,\n  \"temperature\": 0.7,\n  \"top_p\": 0.95,\n  \"tools\": [\n    {\n      \"type\": \"function\",\n      \"function\": {}\n    }\n  ]\n}")

	req, _ := http.NewRequest("POST", url, payload)

	req.Header.Add("Authorization", "Bearer <token>")
	req.Header.Add("Content-Type", "application/json")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

HttpResponse<String> response = Unirest.post("https://direct.evolink.ai/v1/chat/completions")
  .header("Authorization", "Bearer <token>")
  .header("Content-Type", "application/json")
  .body("{\n  \"model\": \"grok-4.5\",\n  \"messages\": [\n    {\n      \"role\": \"system\",\n      \"content\": \"You are a concise assistant.\"\n    },\n    {\n      \"role\": \"user\",\n      \"content\": \"Explain prompt caching in one sentence.\"\n    }\n  ],\n  \"stream\": false,\n  \"max_tokens\": 1024,\n  \"temperature\": 0.7,\n  \"top_p\": 0.95,\n  \"tools\": [\n    {\n      \"type\": \"function\",\n      \"function\": {}\n    }\n  ]\n}")
  .asString();

require 'uri'
require 'net/http'

url = URI("https://direct.evolink.ai/v1/chat/completions")

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

request = Net::HTTP::Post.new(url)
request["Authorization"] = 'Bearer <token>'
request["Content-Type"] = 'application/json'
request.body = "{\n  \"model\": \"grok-4.5\",\n  \"messages\": [\n    {\n      \"role\": \"system\",\n      \"content\": \"You are a concise assistant.\"\n    },\n    {\n      \"role\": \"user\",\n      \"content\": \"Explain prompt caching in one sentence.\"\n    }\n  ],\n  \"stream\": false,\n  \"max_tokens\": 1024,\n  \"temperature\": 0.7,\n  \"top_p\": 0.95,\n  \"tools\": [\n    {\n      \"type\": \"function\",\n      \"function\": {}\n    }\n  ]\n}"

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

{
  "id": "chatcmpl-20260812164515123456789AbCdEfGh",
  "model": "grok-4.5",
  "object": "chat.completion",
  "created": 1786538000,
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Prompt caching reuses previously processed prompt prefixes so repeated context is billed at a lower rate.",
        "tool_calls": [
          {}
        ]
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 504,
    "completion_tokens": 2,
    "total_tokens": 526,
    "prompt_tokens_details": {
      "cached_tokens": 0
    }
  }
}

{
  "error": {
    "code": 400,
    "message": "Invalid request parameters",
    "type": "invalid_request_error"
  }
}

{
  "error": {
    "code": 401,
    "message": "Invalid or expired token",
    "type": "authentication_error"
  }
}

{
  "error": {
    "code": 402,
    "message": "Insufficient quota",
    "type": "insufficient_quota_error",
    "fallback_suggestion": "https://evolink.ai/dashboard/billing"
  }
}

{
  "error": {
    "code": 429,
    "message": "Rate limit exceeded",
    "type": "rate_limit_error",
    "fallback_suggestion": "retry after 60 seconds"
  }
}

{
  "error": {
    "code": 500,
    "message": "Internal server error",
    "type": "internal_server_error",
    "fallback_suggestion": "try again later"
  }
}

{
  "error": {
    "code": 503,
    "message": "Service temporarily unavailable",
    "type": "service_unavailable_error",
    "fallback_suggestion": "retry after 30 seconds"
  }
}

Chat Completions API

Grok Schnittstelle für alle Modelle - Chat Completions vollständige Parameter

OpenAI-kompatibler Chat-Completions-Endpunkt für xAI-Grok-Textmodelle; Modellauswahl über den Parameter model (alle Werte siehe Tabelle beim Parameter model)
grok-4.5: Kontextfenster von 500K Token; ab 200K Token im Prompt werden alle Token-Typen zum doppelten Preis abgerechnet
Prompt-Caching greift automatisch: Token aus dem Cache werden zum günstigeren Cache-Eingabepreis abgerechnet
Synchroner und Streaming-Modus (SSE)
Reguläre function-Tool-Aufrufe werden unterstützt; serverseitige xAI-Tools stehen ausschließlich über die Responses API bereit

POST

chat

completions

Grok Chat-Vervollständigung (alle Modelle, vollständige Parameter)

curl --request POST \
  --url https://direct.evolink.ai/v1/chat/completions \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "grok-4.5",
  "messages": [
    {
      "role": "system",
      "content": "You are a concise assistant."
    },
    {
      "role": "user",
      "content": "Explain prompt caching in one sentence."
    }
  ],
  "stream": false,
  "max_tokens": 1024,
  "temperature": 0.7,
  "top_p": 0.95,
  "tools": [
    {
      "type": "function",
      "function": {}
    }
  ]
}
'

import requests

url = "https://direct.evolink.ai/v1/chat/completions"

payload = {
    "model": "grok-4.5",
    "messages": [
        {
            "role": "system",
            "content": "You are a concise assistant."
        },
        {
            "role": "user",
            "content": "Explain prompt caching in one sentence."
        }
    ],
    "stream": False,
    "max_tokens": 1024,
    "temperature": 0.7,
    "top_p": 0.95,
    "tools": [
        {
            "type": "function",
            "function": {}
        }
    ]
}
headers = {
    "Authorization": "Bearer <token>",
    "Content-Type": "application/json"
}

response = requests.post(url, json=payload, headers=headers)

print(response.text)

const options = {
  method: 'POST',
  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
  body: JSON.stringify({
    model: 'grok-4.5',
    messages: [
      {role: 'system', content: 'You are a concise assistant.'},
      {role: 'user', content: 'Explain prompt caching in one sentence.'}
    ],
    stream: false,
    max_tokens: 1024,
    temperature: 0.7,
    top_p: 0.95,
    tools: [{type: 'function', function: {}}]
  })
};

fetch('https://direct.evolink.ai/v1/chat/completions', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

<?php

$curl = curl_init();

curl_setopt_array($curl, [
  CURLOPT_URL => "https://direct.evolink.ai/v1/chat/completions",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 10,
  CURLOPT_TIMEOUT => 30,
  CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
  CURLOPT_CUSTOMREQUEST => "POST",
  CURLOPT_POSTFIELDS => json_encode([
    'model' => 'grok-4.5',
    'messages' => [
        [
                'role' => 'system',
                'content' => 'You are a concise assistant.'
        ],
        [
                'role' => 'user',
                'content' => 'Explain prompt caching in one sentence.'
        ]
    ],
    'stream' => false,
    'max_tokens' => 1024,
    'temperature' => 0.7,
    'top_p' => 0.95,
    'tools' => [
        [
                'type' => 'function',
                'function' => [
                                
                ]
        ]
    ]
  ]),
  CURLOPT_HTTPHEADER => [
    "Authorization: Bearer <token>",
    "Content-Type: application/json"
  ],
]);

$response = curl_exec($curl);
$err = curl_error($curl);

curl_close($curl);

if ($err) {
  echo "cURL Error #:" . $err;
} else {
  echo $response;
}

package main

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

func main() {

	url := "https://direct.evolink.ai/v1/chat/completions"

	payload := strings.NewReader("{\n  \"model\": \"grok-4.5\",\n  \"messages\": [\n    {\n      \"role\": \"system\",\n      \"content\": \"You are a concise assistant.\"\n    },\n    {\n      \"role\": \"user\",\n      \"content\": \"Explain prompt caching in one sentence.\"\n    }\n  ],\n  \"stream\": false,\n  \"max_tokens\": 1024,\n  \"temperature\": 0.7,\n  \"top_p\": 0.95,\n  \"tools\": [\n    {\n      \"type\": \"function\",\n      \"function\": {}\n    }\n  ]\n}")

	req, _ := http.NewRequest("POST", url, payload)

	req.Header.Add("Authorization", "Bearer <token>")
	req.Header.Add("Content-Type", "application/json")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

HttpResponse<String> response = Unirest.post("https://direct.evolink.ai/v1/chat/completions")
  .header("Authorization", "Bearer <token>")
  .header("Content-Type", "application/json")
  .body("{\n  \"model\": \"grok-4.5\",\n  \"messages\": [\n    {\n      \"role\": \"system\",\n      \"content\": \"You are a concise assistant.\"\n    },\n    {\n      \"role\": \"user\",\n      \"content\": \"Explain prompt caching in one sentence.\"\n    }\n  ],\n  \"stream\": false,\n  \"max_tokens\": 1024,\n  \"temperature\": 0.7,\n  \"top_p\": 0.95,\n  \"tools\": [\n    {\n      \"type\": \"function\",\n      \"function\": {}\n    }\n  ]\n}")
  .asString();

require 'uri'
require 'net/http'

url = URI("https://direct.evolink.ai/v1/chat/completions")

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

request = Net::HTTP::Post.new(url)
request["Authorization"] = 'Bearer <token>'
request["Content-Type"] = 'application/json'
request.body = "{\n  \"model\": \"grok-4.5\",\n  \"messages\": [\n    {\n      \"role\": \"system\",\n      \"content\": \"You are a concise assistant.\"\n    },\n    {\n      \"role\": \"user\",\n      \"content\": \"Explain prompt caching in one sentence.\"\n    }\n  ],\n  \"stream\": false,\n  \"max_tokens\": 1024,\n  \"temperature\": 0.7,\n  \"top_p\": 0.95,\n  \"tools\": [\n    {\n      \"type\": \"function\",\n      \"function\": {}\n    }\n  ]\n}"

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

{
  "id": "chatcmpl-20260812164515123456789AbCdEfGh",
  "model": "grok-4.5",
  "object": "chat.completion",
  "created": 1786538000,
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Prompt caching reuses previously processed prompt prefixes so repeated context is billed at a lower rate.",
        "tool_calls": [
          {}
        ]
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 504,
    "completion_tokens": 2,
    "total_tokens": 526,
    "prompt_tokens_details": {
      "cached_tokens": 0
    }
  }
}

{
  "error": {
    "code": 400,
    "message": "Invalid request parameters",
    "type": "invalid_request_error"
  }
}

{
  "error": {
    "code": 401,
    "message": "Invalid or expired token",
    "type": "authentication_error"
  }
}

{
  "error": {
    "code": 402,
    "message": "Insufficient quota",
    "type": "insufficient_quota_error",
    "fallback_suggestion": "https://evolink.ai/dashboard/billing"
  }
}

{
  "error": {
    "code": 429,
    "message": "Rate limit exceeded",
    "type": "rate_limit_error",
    "fallback_suggestion": "retry after 60 seconds"
  }
}

{
  "error": {
    "code": 500,
    "message": "Internal server error",
    "type": "internal_server_error",
    "fallback_suggestion": "try again later"
  }
}

{
  "error": {
    "code": 503,
    "message": "Service temporarily unavailable",
    "type": "service_unavailable_error",
    "fallback_suggestion": "retry after 30 seconds"
  }
}

BaseURL: Die Standard-BaseURL ist https://direct.evolink.ai und bietet bessere Unterstützung für Textmodelle sowie persistente Verbindungen. https://api.evolink.ai ist der primäre Endpunkt für multimodale Dienste und dient bei Textmodellen als Ausweichadresse.

Serverseitige Tools (Websuche, X-Suche, Codeausführung, Anhangsuche, Sammlungssuche) stehen ausschließlich über die Responses API bereit. Der Chat-Completions-Endpunkt unterstützt nur reguläre function-Tool-Aufrufe.

Autorisierungen

Authorization

string

header

erforderlich

##Alle APIs erfordern Bearer-Token-Authentifizierung##

API-Schlüssel erhalten:

Besuchen Sie die API-Schlüsselverwaltungsseite, um Ihren API-Schlüssel zu erhalten

Zum Anfrage-Header hinzufügen:

Authorization: Bearer YOUR_API_KEY

Body

application/json

model

enum<string>

erforderlich

Aufzurufendes Modell:

Modell-ID	Positionierung
`grok-4.5`	xAI-Modell für Reasoning und Tool-Aufrufe, Kontextfenster von 500K

Verfügbare Optionen:

grok-4.5

Beispiel:

"grok-4.5"

messages

object[]

erforderlich

Liste der Chat-Nachrichten. Unterstützt die Rollen system, user und assistant.

Minimum array length: 1

Show child attributes

Beispiel:

[
  {
    "role": "system",
    "content": "You are a concise assistant."
  },
  {
    "role": "user",
    "content": "Explain prompt caching in one sentence."
  }
]

stream

boolean

Standard:false

Ob eine Streaming-Antwort zurückgegeben wird (SSE, chat.completion.chunk-Events). Standard false.

Beispiel:

false

max_tokens

integer

Maximale Anzahl der zu generierenden Token. Wird unverändert an das Modell weitergegeben.

Beispiel:

1024

temperature

number

Sampling-Temperatur (0-2). Höhere Werte erzeugen zufälligere Ausgaben.

Beispiel:

0.7

top_p

number

Nucleus-Sampling-Parameter (0-1).

Beispiel:

0.95

tools

object[]

Reguläre OpenAI-function-Tool-Definitionen (clientseitige Funktionsaufrufe, keine zusätzliche Gebühr pro Aufruf). Serverseitige xAI-Tools stehen ausschließlich über die Responses API bereit.

Show child attributes

tool_choice

Steuert die Funktionsauswahl: "auto" / "none" / "required" oder ein Objekt, das eine bestimmte Funktion festlegt.

Verfügbare Optionen:

auto,

none,

required

Antwort

Chat-Vervollständigung erfolgreich generiert (JSON-Objekt oder – bei stream=true – ein SSE-Stream von chat.completion.chunk-Events)

string

Eindeutiger Bezeichner für die Chat-Vervollständigung

Beispiel:

"chatcmpl-20260812164515123456789AbCdEfGh"

model

string

Tatsächlich verwendeter Modellname

Beispiel:

"grok-4.5"

object

enum<string>

Antworttyp

Verfügbare Optionen:

chat.completion

Beispiel:

"chat.completion"

created

integer

Erstellungszeitstempel

Beispiel:

1786538000

choices

object[]

Liste der Chat-Vervollständigungsoptionen

Show child attributes

usage

object

Statistik zur Token-Nutzung. Ab 200K Token im Prompt werden alle Token-Typen (Eingabe, Cache-Eingabe, Ausgabe) zum doppelten Preis abgerechnet.

Show child attributes

Bildserie

Videoserie

Audioserie

Textserie

Kontoverwaltung

Aufgabenverwaltung

Dateiverwaltung

Grok Schnittstelle für alle Modelle - Chat Completions vollständige Parameter

Autorisierungen

Body

Antwort