Model by xAI

Grok 4.6

Grok 4.6 is xAI's model for long-running coding agents, and it suits teams that already use Grok Build or Cursor.

Checked 2026-11-05

Grok 4.6 is xAI's model for long-running coding agents, and it suits teams that already use Grok Build or Cursor. The maker's launch page, dated 12 August 2026, calls it the successor to Grok 4.5, aimed at research, codebase work and turning an idea into a working app. The docs give a 500,000 token context window, text and image input with text output, and reasoning effort from low to xhigh. Price is $2 per million input tokens and $6 per million output tokens for prompts under 200,000 tokens. At 200,000 or more, the whole request is billed at $4 and $12. Batch requests get 20 percent off. The page says it ties GPT-5.6 Sol on the Artificial Analysis index at 61, but its own table shows it trailing Sol on DeepSWE and Terminal-Bench. OpenRouter notes that Grok 4.7 has since succeeded it, and the xAI docs now recommend 4.7 for code and chat. You can reach it through the xAI API, Cursor, Grok Build, OpenRouter, Vercel and Cloudflare. We would use it only on existing integrations and test Grok 4.7 for new ones. Figures come from xAI's launch page, its docs and OpenRouter. We have not run it ourselves.

Mentioned here

  • Grok 4.7

    Grok 4.7 is a closed xAI model, released 2026-09-21, listed on OpenRouter at $2 per million input tokens and $6 output with a 500,000-token context and text, image and PDF input.

    • Closed

    xAI, 500K context, $2 in / 1M

  • Grok 4.5

    Grok 4.5 is xAI's earlier coding-focused model, and it suits teams that still have it set as the default in Grok Build or Cursor.

    • Closed

    xAI, 500K context, $2 in / 1M

  • GPT-5.6 Sol

    GPT-5.6 Sol is OpenAI's flagship in the 5.6 series, and it suits developers who want strong command-line and multi-step coding work through the API.

    • Closed

    OpenAI, 1.1M context, $4 in / 1M

  • xAI API

    xAI's Grok API: grok-4.7 $2 in, $6 out with a 500K context. Spend tiers, 30-day audit storage, zero data retention switch, Grok Build and agent setup.

    • Model / inference API
    • OpenAI compatible

  • OpenRouter

    OpenRouter routes more than 600 models through one API key.

    • Model gateway (router)
    • OpenAI compatible

  • Grok Build

    Grok Build is xAI's terminal coding agent with plan mode, subagents and MCP. Free to try; Plus $8, Premium $30, Super $100, Ultra $200 a month.

    • Terminal agent
    • MCP

    Subscription

Questions

What is the context window of Grok 4.6?

500,000 tokens, with up to 450,000 tokens of output.

How much does Grok 4.6 cost?

$2 per million input tokens and $6 per million output tokens, checked 31 October 2026.

Can I run Grok 4.6 on my own hardware?

No. The weights are not published, so you use it through the maker's API or a hosting provider.

Who makes Grok 4.6?

xAI makes it, and released it on 12 August 2026.

Reviews

Loading reviews…

Write a review

Your rating

Up to 1000 characters. Stars alone are enough.

Similar

  • Grok 4.3

    Grok 4.3 is a reasoning model from xAI, now listed by OpenRouter under the name SpaceXAI, that suits people who want a one million token context at a low output price.

    • Closed

    xAI, 1M context, $1.25 in / 1M