grok-4.7

Grok 4.7 is xAI’s frontier reasoning model for software development, agent execution, and professional knowledge work. It is designed to improve long-horizon task execution, complex planning, code debugging, and self-verification. It is well suited for software engineering, research and analysis, document and presentation creation, and automated workflows that require multiple rounds of tool use. The model supports text and image input with text output, offers a 500,000-token context window, and has no text output limit. Developers can choose among low, medium, high, and xhigh reasoning effort levels to balance speed, cost, and reasoning quality. It supports function calling, web search, X search, and code execution, and can be accessed through xAI’s Responses API or Chat Completions API.

Context

500000 tokens

Max output

128000 tokens

Input

llm image

Output

llm