MODELS

xAI: Grok 4.20 Multi-Agent

by xAI

modellead-sourceopenrouter-modelsxaigrokmulti-agentresearchtext-inputimage-inputsource:x.aigrok-4.20reasoningapi

Overview

Grok 4.20 Multi-Agent Beta is an xAI API model for text/image-input, large-context, multi-agent research workflows.

Details

xAI documentation lists Grok 4.20 Multi-Agent Beta with model name grok-4.20-multi-agent-0309 and alias grok-4.20-multi-agent. The supplied model metadata says it supports text and image input, has a 1,000,000-token context window, and is priced per 1M input and output tokens. xAI’s multi-agent documentation describes Realtime Multi-agent Research, where Grok orchestrates multiple agents for deep, multi-step research, and instructs developers to use grok-4.20-multi-agent in API requests. The Grok 4.20 system card states that Grok 4.20 supports text and image inputs with text outputs and has advanced reasoning and multi-agent capabilities.

When to Use

Use for deep, multi-step research tasks where xAI’s Realtime Multi-agent Research orchestration is appropriate. Use when an API workflow needs text and image input support with a documented 1,000,000-token context window. Use when evaluating Grok 4.20 variants for reasoning-oriented API workflows.

Getting Started

Review the official xAI model page for grok-4.20-multi-agent-0309.
Read xAI’s multi-agent guide for how to call grok-4.20-multi-agent in API requests.
Check the official xAI models and pricing page before production use.
Run a small API test on representative research prompts and verify outputs for your application.

Key Features

•Model name: grok-4.20-multi-agent-0309
•Alias: grok-4.20-multi-agent
•Text and image input support
•1,000,000-token context window
•Realtime Multi-agent Research orchestration for deep, multi-step research
•API pricing listed at $1.25 per 1M input tokens and $2.50 per 1M output tokens

Capabilities

•text-input
•image-input
•text-output
•multi-agent-research
•large-context
•api-model

Last updated Jun 2, 2026