Grok 4.7

xai/grok-4.7

LLMReasoningCodingAgenticTool UseMultimodal

xAI's Grok 4.7, a frontier reasoning model for coding, agentic tasks, and knowledge work. Accepts text and image inputs, and supports function calling, structured outputs, and configurable reasoning effort.

Quick start

# Inspect the price — a plain request returns the 402 challenge:
curl -i https://api.glianalabs.com/v1/chat/completions \
  -H "content-type: application/json" \
  -d '{
    "model": "xai/grok-4.7",
    "messages": [{"role": "user", "content": "<your message>"}]
  }'

# Pay + run in one step with the mppx CLI (create a wallet: npx mppx account create):
npx mppx https://api.glianalabs.com/v1/chat/completions \
  -J '{"model": "xai/grok-4.7", "messages": [{"role": "user", "content": "<your message>"}]}'

Parameters

Input

messages array required

max_completion_tokens integer optional

max_tokens integer optional

n integer optional

How many outputs to generate.

parallel_tool_calls boolean optional

prompt_cache_key string optional

reasoning_effort string optional

response_format object optional

How the result is returned (e.g. a URL or base64 data).

search_parameters object optional

seed integer optional

Random seed for reproducible results — same seed + inputs gives the same output.

service_tier string optional

stream boolean optional

stream_options object optional

temperature number optional

tool_choice string optional

tools array optional

top_p number optional

user string optional

Optional end-user identifier for abuse monitoring; not required.

web_search_options object optional

Output