Completado de chat Grok (todos los modelos, parámetros completos)

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

Interfaz de todos los modelos Grok - Referencia completa de Chat Completions

Endpoint Chat Completions compatible con OpenAI para los modelos de texto xAI Grok; selecciona el modelo con el parámetro model (todos los valores están en la tabla del parámetro model)
grok-4.5: ventana de contexto de 500K tokens; a partir de 200K tokens en el prompt, todos los tipos de token se facturan al doble
El almacenamiento en caché del prompt es automático: los tokens de entrada servidos desde la caché se facturan a la tarifa de caché, más baja
Modos sincrónico y streaming (SSE)
Se admiten las llamadas a herramientas function normales; las herramientas del lado del servidor de xAI solo están disponibles en la Responses API

POST

chat

completions

Completado de chat Grok (todos los modelos, parámetros completos)

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: La BaseURL predeterminada es https://direct.evolink.ai, que ofrece mejor compatibilidad con modelos de texto y admite conexiones persistentes. https://api.evolink.ai es el endpoint principal para servicios multimodales y actúa como dirección de respaldo para los modelos de texto.

Las herramientas del lado del servidor (búsqueda web, búsqueda en X, ejecución de código, búsqueda de adjuntos, búsqueda de colecciones) solo están disponibles en la Responses API. El endpoint Chat Completions solo admite las llamadas a herramientas function normales.

Autorizaciones

Authorization

string

header

requerido

##Todas las APIs requieren autenticación Bearer Token##

Obtener API Key:

Visita la Página de gestión de API Key para obtener tu API Key

Agregar al encabezado de la solicitud:

Authorization: Bearer YOUR_API_KEY

Cuerpo

application/json

model

enum<string>

requerido

Modelo a invocar:

ID del modelo	Posicionamiento
`grok-4.5`	Modelo xAI de razonamiento y uso de herramientas, ventana de contexto de 500K

Opciones disponibles:

grok-4.5

Ejemplo:

"grok-4.5"

messages

object[]

requerido

Lista de mensajes de chat. Admite los roles system, user y assistant.

Minimum array length: 1

Show child attributes

Ejemplo:

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

stream

boolean

predeterminado:false

Indica si se devuelve una respuesta en streaming (SSE, eventos chat.completion.chunk). Predeterminado false.

Ejemplo:

false

max_tokens

integer

Número máximo de tokens a generar. Se transmite al modelo tal cual.

Ejemplo:

1024

temperature

number

Temperatura de muestreo (0-2). Los valores más altos producen una salida más aleatoria.

Ejemplo:

0.7

top_p

number

Parámetro de muestreo nucleus (0-1).

Ejemplo:

0.95

tools

object[]

Definiciones de herramientas function normales de OpenAI (llamadas a funciones del lado del cliente, sin coste adicional por llamada). Las herramientas del lado del servidor de xAI solo están disponibles en la Responses API.

Show child attributes

tool_choice

Controla la selección de la función: "auto" / "none" / "required", o un objeto que fija una función concreta.

Opciones disponibles:

auto,

none,

required

Respuesta

Completado de chat generado exitosamente (objeto JSON, o un flujo SSE de eventos chat.completion.chunk cuando stream=true)

string

Identificador único para la completación de chat

Ejemplo:

"chatcmpl-20260812164515123456789AbCdEfGh"

model

string

Nombre del modelo realmente utilizado

Ejemplo:

"grok-4.5"

object

enum<string>

Tipo de respuesta

Opciones disponibles:

chat.completion

Ejemplo:

"chat.completion"

created

integer

Marca de tiempo de creación

Ejemplo:

1786538000

choices

object[]

Lista de opciones de completado de chat

Show child attributes

usage

object

Estadísticas de uso de tokens. Los prompts de 200K tokens o más se facturan al doble en todos los tipos de token (entrada, entrada en caché, salida).

Show child attributes

Serie de imágenes

Serie de videos

Serie de audio

Serie de texto

Gestión de cuenta

Gestión de tareas

Gestión de archivos

Interfaz de todos los modelos Grok - Referencia completa de Chat Completions

Autorizaciones

Cuerpo

Respuesta