Category

API

App

Grok
Date
Price

grok-4-1-fast-non-reasoning

The non-inference mode of Grok-4-1-Fast, designed for real-time response scenarios that require extremely low latency.
Model
LLM
Model capability: imageModel capability: function_call
Input:
$0.2/1M tokens
Output:
$0.5/1M tokens

grok-4-1-fast-reasoning

A cutting-edge multimodal model optimized specifically for high-performance agent tool calls, designed for high-complexity reasoning and large-scale agent systems.
Model
LLM
Model capability: imageModel capability: thinkingModel capability: function_call
Input:
$0.2/1M tokens
Output:
$0.5/1M tokens

grok-4-fast-reasoning

Grok-4-Fast’s deep reasoning mode, used for step-by-step chain-of-thought analysis.
Model
LLM
Model capability: imageModel capability: thinkingModel capability: function_call
Input:
$0.2/1M tokens
Output:
$0.5/1M tokens

grok-code-fast-1

A speedy and economical reasoning model that excels at agentic coding
Model
LLM
Model capability: thinkingModel capability: function_call
Input:
$0.2/1M tokens
Output:
$1.5/1M tokens

grok-4-0709

An intelligent model launched by xAI, featuring native tool integration, real-time search capabilities, and cutting-edge reasoning abilities.
Model
LLM
Model capability: function_call
Input:
$3/1M tokens
Output:
$15/1M tokens

grok-3-mini

A streamlined version based on the Grok-3 core dataset, featuring fast inference speed and low latency, suitable for everyday lightweight scenarios.
Model
LLM
Input:
$0.3/1M tokens
Output:
$0.5/1M tokens

grok-3

The most advanced AI model series launched by xAI, the “Reasoning Agent,” which combines exceptional reasoning capabilities with vast pre-trained knowledge.
Model
LLM
Input:
$3/1M tokens
Output:
$15/1M tokens

grok-2-vision-1212

Grok-2’s vision model achieves state-of-the-art performance in both visual mathematical reasoning and document-based question answering.
Model
LLM
Model capability: imageModel capability: function_call
Input:
$2/1M tokens
Output:
$10/1M tokens