# Grok 4.20 Multi Agent Beta

The beta release of SpaceXAI's Grok 4.20 multi-agent variant. Grok 4.20 Multi Agent Beta provides early access to agent orchestration capabilities, with the beta tag indicating the model may change before stabilization.

- **Model ID:** `spacexai/grok-4.20-multi-agent-beta`
- **Type:** chat
- **Providers:** xai
- **Context window:** 2,000,000
- **Maximum output tokens:** 2,000,000
- **Pricing:** $1.25/1M input tokens, $2.5/1M output tokens
- **Canonical page:** https://vercel.com/ai-gateway/models/grok-4.20-multi-agent-beta

## Supported parameters

Detailed capability metadata has not been reported for this model.

## Example

```ts
import { streamText } from 'ai'

const result = streamText({
  model: 'spacexai/grok-4.20-multi-agent-beta',
  prompt: 'Why is the sky blue?'
})
```

## About

Grok 4.20 Multi Agent Beta was released March 11, 2026 as the beta-tagged variant of SpaceXAI's Grok 4.20 multi-agent model. The beta designation means SpaceXAI may still update the model's weights, orchestration behavior, or agent coordination patterns before declaring a stable release.

Grok 4.20 Multi Agent Beta shares the same multi-agent orchestration architecture as its stable counterpart: coordination between multiple agents, task delegation, and collaborative workflows. Multi-agent orchestration is still an emerging pattern, and the beta lifecycle gives SpaceXAI room to refine how agents discover capabilities, negotiate task boundaries, and handle failures.

Both the beta and stable multi-agent variants are available through Vercel AI Gateway under separate identifiers. Teams building multi-agent systems can experiment with Grok 4.20 Multi Agent Beta while keeping the non-beta grok-4.20-multi-agent as the production fallback.

## What to consider

Grok 4.20 Multi Agent Beta is in beta. Expect potential changes before general availability.

This model is optimized for agent collaboration patterns. For single-agent workflows, the non-reasoning or reasoning Grok 4.20 variants may be more appropriate.

## When to use

### Best For

- **Multi-agent orchestration** where several agents coordinate on complex tasks
- **Agent delegation workflows** where a supervisor agent routes subtasks to specialist agents
- **Complex automation pipelines** requiring multiple agents with different roles to collaborate
- **Tool-calling coordination** where agents need to share context and hand off results reliably
- **Research and analysis systems** that decompose large problems across specialized agent roles

### Consider Alternatives When

- **Single-agent workflows** where Grok 4.20 Non-Reasoning or Reasoning variants are better suited
- **Speed-focused single-agent tasks** where Grok 4.1 Fast Non-Reasoning provides fast tool calling without multi-agent overhead
- **Stable production deployments** where beta models introduce risk. Wait for general availability or use Grok 4.1 Fast models
- **Simple text generation** where general-purpose Grok models handle the task without multi-agent optimization

## Best for

- **Multi-agent orchestration:** workflows where several agents coordinate tool calls, share context, and hand off subtasks reliably
- **Agent delegation workflows:** a supervisor agent routes subtasks to specialist agents
- **Complex automation pipelines** requiring multiple agents with different roles to collaborate
- **Tool-calling coordination:** agents need to share context and hand off results reliably
- **Research and analysis systems** that decompose large problems across specialized agent roles

## Consider alternatives

- **Single-agent workflows:** Grok 4.20 Non-Reasoning or Reasoning variants are better suited
- **Speed-focused single-agent tasks:** Grok 4.1 Fast Non-Reasoning provides fast tool calling without multi-agent overhead
- **Stable production deployments:** beta models introduce risk. Wait for general availability or use Grok 4.1 Fast models
- **Simple text generation:** general-purpose Grok models handle the task without multi-agent optimization

## Frequently asked questions

### What makes Grok 4.20 Multi Agent Beta different from other Grok 4.20 models?

Grok 4.20 Multi Agent Beta is purpose-built for multi-agent orchestration. While the reasoning and non-reasoning variants target individual agent tasks, this model is optimized for workflows where multiple agents coordinate and collaborate.

### What is the context window for Grok 4.20 Multi Agent Beta?

2M tokens.

### Is Grok 4.20 Multi Agent Beta in beta?

Yes. Grok 4.20 Multi Agent Beta is currently in beta. Behavior, pricing, or availability may change before general availability.

### What does Grok 4.20 Multi Agent Beta cost?

Check the pricing panel on this page for today's numbers. AI Gateway tracks rates across every provider that serves Grok 4.20 Multi Agent Beta.

### How do I authenticate with Grok 4.20 Multi Agent Beta through Vercel AI Gateway?

Use your Vercel AI Gateway API key with `spacexai/grok-4.20-multi-agent-beta` as the model identifier. No separate SpaceXAI account is needed for gateway-managed access.

### Can I use Grok 4.20 Multi Agent Beta for single-agent workflows?

You can, but it's optimized for multi-agent patterns. For single-agent use cases, the Grok 4.20 Non-Reasoning or Reasoning variants are better choices.

### Does Vercel AI Gateway support Zero Data Retention for Grok 4.20 Multi Agent Beta?

Yes, Zero Data Retention is available for this model. Zero Data Retention is offered on a per-provider basis. See https://vercel.com/docs/ai-gateway/capabilities/zdr for details.

## Links

- [Model page](https://vercel.com/ai-gateway/models/grok-4.20-multi-agent-beta)
- [AI Gateway documentation](https://vercel.com/docs/ai-gateway)
- [Provider model documentation](https://docs.x.ai/developers/models/grok-4.20-multi-agent-beta-0309)
- [Provider pricing](https://docs.x.ai/developers/models)