New Docs Section: Grok API Calls

2026/7/13 10:49 (UTC+8) · Docs Update · xAI 📚 New docs section “Grok API Calls” — 5 hands-on-verified bilingual pages now live We ran a full hands-on test of all 6 Grok models (grok-4.5 / grok-4.3 / both grok-4.20 variants / grok-build-0.1 / multi-agent) on APIYI — 56 request logs, every capability boundary verified — and turned the results into documentation under “Basic API → Grok API Calls”:

Grok Overview: model lineup, pricing (matching xAI official), verified capability matrix, and the unsupported-features list
Chat & Reasoning: streaming, chain-of-thought billing essentials, structured outputs, function calling, vision input
Web & X Search: web_search / x_search verified performing real searches; X search is unique to Grok
Code Execution & MCP: server-side Python sandbox and Remote MCP verified working
Multi-Agent Model: capabilities and the billing-amplification profile of multi-agent collaboration

Feedback from real-world usage is always welcome — we keep the docs improving.

← Back to Live Updates · 📚 Monthly Archive