Grok 4 Fast Reasoning

by xAI · chat

Reasoning-enabled variant of Grok 4 Fast, a cost-efficient multimodal model with 2M context window. Achieves comparable performance to Grok 4 while using approximately 40%% fewer thinking tokens.

Pricing

Input: $0.19/M tokens · Output: $0.475/M tokens

Capabilities

Vision, Function Calling, JSON Mode, Streaming, Reasoning

Context: 2000K tokens

Max output: 16K tokens

Routes: 2/2 healthy

Performance

TTFT: 1361ms · Latency: 3947ms · Throughput: 170.8 tok/s