Enable image understanding in web search

Set enable_image_understanding to true on the web_search tool. Grok gets the view_image tool and can analyze images it encounters while browsing pages. Usage shows up as SERVER_SIDE_TOOL_VIEW_IMAGE in response.server_side_tool_usage. Base setup: Add web search to a Grok API request. To embed image search results as Markdown instead, see Enable image search in web search.

xAI SDK

import os

from xai_sdk import Client
from xai_sdk.chat import user
from xai_sdk.tools import web_search

client = Client(api_key=os.getenv("XAI_API_KEY"))
chat = client.chat.create(
    model="grok-4.6",
    tools=[
        web_search(enable_image_understanding=True),
    ],
)

chat.append(user("What is included in the image on xAI's official website?"))
response = chat.sample()
print(response.content)
print(response.server_side_tool_usage)

OpenAI Responses

response = client.responses.create(
    model="grok-4.6",
    input=[
        {
            "role": "user",
            "content": "What is included in the image on xAI's official website?",
        },
    ],
    tools=[
        {
            "type": "web_search",
            "enable_image_understanding": True,
        },
    ],
)

print(response)

Vercel AI SDK

import { xai } from '@ai-sdk/xai';
import { generateText } from 'ai';

const { text } = await generateText({
  model: xai.responses('grok-4.6'),
  prompt: "What is included in the image on xAI's official website?",
  tools: {
    web_search: xai.tools.webSearch({
      enableImageUnderstanding: true,
    }),
  },
});

console.log(text);

Pitfalls

• Enabling this on Web Search also enables image understanding for X Search when that tool is on the same request.
• enable_image_search is a different flag: it finds images and embeds them as Markdown. This flag inspects images on pages already being browsed.
• Console API credits are separate from SuperGrok's weekly pool on grok.com.