Grok 4 Fast Non-Reasoning

JSON →
xai llm
text

A fast inference variant of Grok 4 without reasoning capabilities for speed.

context window 2.0M tokens
max output 2.0M tokens
input price $0.2 / 1M tokens
output price $0.5 / 1M tokens
streamingcode-generationfunction-callingtool-useprompt-caching