Grok 4.20 Beta

xAI

Latest Grok model with 2M context window and fastest output speed of any tracked model (265 tok/s). Available in reasoning and non-reasoning variants.

Text Generation API Grok Family v4.20

Documentation Back to Models

Parameters

params

Context Window

2.0M

tokens

Max Output

tokens

Input Price

per 1M tokens

Output Price

per 1M tokens

Grok Family 9 models

The full Grok line by generation — pricing and capabilities vary across the family.

4.3

4.3 Latest

General reasoning, coding, multimodal (video) understanding

$1.25 / $2.5

in / out · 1M

Apr 2026

4.20

4.20 Beta Current

General-purpose, reasoning, fast inference

$2 / $6

in / out · 1M

Mar 2026

Grok

Real-time knowledge, wit

Real-time knowledge, humor

$5 / $15

in / out · 1M

Feb 2025

Capabilities

👁️

Vision

⚡

Function Calling

📋

JSON Mode

🌊

Streaming

💬

System Prompt

🖥️

Code Execution

🔍

Web Search

🔌

MCP Support

Extended Pricing

Input

$2/1M

Output

$6/1M

Cache Read

$0.2/1M

Details

Release Date: March 10, 2026
Knowledge Cutoff: November 1, 2024
Source: API
API Key Required: Yes
License: Proprietary
Model ID: grok-4.20-beta

Last updated: May 27, 2026