xAI

Grok 4.20 Non Reasoning

Grok 4.20 Non Reasoning is a large language model by xAI with support for vision, function calling, and structured output. It features an extended 2M token context window, suitable for processing long documents and complex conversations. Priced at $1.25 per million input tokens and $2.5 per million output tokens. Access Grok 4.20 Non Reasoning through Inworld Router with OpenAI SDK compatibility, built-in failover, and intelligent routing across providers.

→Function CallingResponse SchemaVisionresponseFormatCapability

Use This Model

Step 1 — Create Router

curl --location 'https://api.inworld.ai/router/v1/routers' \
--header 'Authorization: Basic <your-api-key>' \
--header 'Content-Type: application/json' \
--data '{
  "name": "compare-frontier-models",
  "defaultRoute": {
    "route_id": "default",
    "variants": [
      {
        "variant": {
          "variant_id": "xai/grok-4.20-non-reasoning-a",
          "model_id": "xai/grok-4.20-non-reasoning"
        },
        "weight": 33
      },
      {
        "variant": {
          "variant_id": "xai/grok-4.20-b",
          "model_id": "xai/grok-4.20"
        },
        "weight": 33
      },
      {
        "variant": {
          "variant_id": "xai/grok-4.20-non-reasoning-c",
          "model_id": "xai/grok-4.20-non-reasoning"
        },
        "weight": 34
      }
    ]
  }
}'

import requests

response = requests.post(
    "https://api.inworld.ai/router/v1/routers",
    headers={
        "Authorization": "Basic <your-api-key>",
        "Content-Type": "application/json",
    },
    json={
        "name": "compare-frontier-models",
        "defaultRoute": {
            "route_id": "default",
            "variants": [
                {
                    "variant": {
                        "variant_id": "xai/grok-4.20-non-reasoning-a",
                        "model_id": "xai/grok-4.20-non-reasoning"
                    },
                    "weight": 33
                },
                {
                    "variant": {
                        "variant_id": "xai/grok-4.20-b",
                        "model_id": "xai/grok-4.20"
                    },
                    "weight": 33
                },
                {
                    "variant": {
                        "variant_id": "xai/grok-4.20-non-reasoning-c",
                        "model_id": "xai/grok-4.20-non-reasoning"
                    },
                    "weight": 34
                }
            ]
        }
    },
)
print(response.json())

const response = await fetch(
  "https://api.inworld.ai/router/v1/routers",
  {
    method: "POST",
    headers: {
      Authorization: "Basic <your-api-key>",
      "Content-Type": "application/json",
    },
    body: JSON.stringify({
      name: "compare-frontier-models",
      defaultRoute: {
        route_id: "default",
        variants: [
          {
            variant: {
              variant_id: "xai/grok-4.20-non-reasoning-a",
              model_id: "xai/grok-4.20-non-reasoning",
            },
            weight: 33,
          },
          {
            variant: {
              variant_id: "xai/grok-4.20-b",
              model_id: "xai/grok-4.20",
            },
            weight: 33,
          },
          {
            variant: {
              variant_id: "xai/grok-4.20-non-reasoning-c",
              model_id: "xai/grok-4.20-non-reasoning",
            },
            weight: 34,
          },
        ],
      },
    }),
  }
);
const data = await response.json();
console.log(data);

Step 2 — Chat Completion

curl --location 'https://api.inworld.ai/v1/chat/completions' \
--header 'Authorization: Basic <your-api-key>' \
--header 'Content-Type: application/json' \
--data '{
  "model": "inworld/compare-frontier-models",
  "messages": [{"role": "user", "content": "Hello!"}]
}'

Grok 4.20 Non Reasoning pricing and providers

Access Grok 4.20 Non Reasoning through xAI via Inworld Router or Realtime API. By using the AI provider(s) below you acknowledge you reviewed and agree to their terms listed in the Legal section under the AI provider's name.