An xAI reasoning model for coding, agentic tasks, and long-context text-and-image analysis.
Use xai/grok-4.7 through the sjolt LLM API.
xai/grok-4.7Token rates
xai/grok-4.7Similar Models
A fast DeepSeek model for text generation and reasoning tasks.
A DeepSeek model for text generation and reasoning tasks that need more capability.
Google’s fast reasoning model for agentic coding and complex knowledge workflows.
Model overview
Connect code, documents, and visual context in a reasoning workflow. The artwork on this page illustrates these capabilities; it is not image output from Grok 4.7.
Describe the goal, provide relevant code and constraints, and use function calls to connect your application tools. Return tool results to continue the reasoning loop.
Bring text and images into a 500,000-token context window. Supply the relevant source material and specify the desired response structure for focused analysis.
Select low, medium, high, xhigh reasoning effort. The Playground defaults to high; use a lighter level for simpler work or xhigh for deeper analysis.
Model characteristics
Coding, agentic tasks, and knowledge work with text and image inputs.
Up to 500,000 context tokens.
Text output with no separately specified output-token ceiling; the context window and upstream constraints still apply.
low, medium, high, xhigh; high is the default in the Playground.
FAQ
Use it for software engineering, multi-step reasoning, document analysis, and workflows that call your application tools. State the goal, constraints, and expected output clearly.
Yes. Send text and image content through Chat Completions or Responses, or attach images in the Playground. Image fields pass through unchanged for upstream validation; responses are text.
Use low, medium, high, xhigh. The Playground defaults to high.
The context window is 500,000 tokens. No separate fixed text-output ceiling is specified; caller-selected token limits are forwarded to the upstream.
Use model xai/grok-4.7 with OpenAI-compatible Chat Completions or Responses through the sjolt LLM API. Both streaming and non-streaming requests use the same model route.
Include the prior response items in the next request input, preserving any reasoning.encrypted_content unchanged alongside the relevant conversation and tool results.