Grok 4.20 Multi-Agent
Grok 4.20 Multi-Agent is xAI's model purpose-built for multi-agent orchestration and collaboration, part of the Grok 4.20 beta generation.
import { streamText } from 'ai'
const result = streamText({ model: 'xai/grok-4.20-multi-agent', prompt: 'Why is the sky blue?'})Playground
Try out Grok 4.20 Multi-Agent by xAI. Usage is billed to your team at API rates. Free users (those who haven't made a payment) get $5 of credits every 30 days.
Providers
Route requests across multiple providers. Copy a provider slug to set your preference. Visit the docs for more info. Using a provider means you agree to their terms, listed under Legal.
P50 throughput on live AI Gateway traffic, in tokens per second (TPS). Visit the docs for more info.
P50 time to first token (TTFT) on live AI Gateway traffic, in milliseconds. View the docs for more info.
Direct request success rate on AI Gateway and per-provider. Visit the docs for more info.
More models by xAI
| Model |
|---|
| 256K | 0.3s | 175tps | $1.00/M | $2.00/M | Read: $0.2/M Write: — | $5/K + input costs | — |  | 05/20/2026 | ||||
| 1M | 1.0s | 102tps | $1.25/M | $2.50/M | Read: $0.2/M Write: — | $5/K + input costs | — | | 04/30/2026 | ||||
| 2M | 0.5s | 177tps | $1.25/M | $2.50/M | Read: $0.2/M Write: — | $5/K + input costs | — | | 03/11/2026 | ||||
| 2M | 0.4s | 81tps | $1.25/M | $2.50/M | Read: $0.2/M Write: — | $5/K + input costs | — |  | 03/09/2026 | ||||
| 1M | 1.0s | 249tps | $0.20/M | $0.50/M | Read:$0.05/M Write:— | — | — | | 07/09/2025 | ||||
| 1M | 0.5s | 66tps | $0.20/M | $0.50/M | Read:$0.05/M Write:— | — | — | | 07/09/2025 |
About Grok 4.20 Multi-Agent
Grok 4.20 Multi-Agent was released March 9, 2026 as part of xAI's Grok 4.20 beta generation, purpose-built for multi-agent orchestration and collaboration. While the other Grok 4.20 variants focus on individual reasoning or speed, Grok 4.20 Multi-Agent targets workflows where multiple agents work together, coordinating tool calls, sharing context, and delegating subtasks.
The model features the same low hallucination rates and strict prompt adherence as its Grok 4.20 siblings, applied specifically to multi-agent challenges: consistent behavior across agent interactions, reliable handoffs, and predictable tool-calling patterns.
This model is currently in beta.
What To Consider When Choosing a Provider
- Configuration: Grok 4.20 Multi-Agent is in beta. Expect potential changes before general availability.
- Configuration: 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.
- Zero Data Retention: AI Gateway does not currently support Zero Data Retention for this model. See the documentation for models that support ZDR.
- Authentication: AI Gateway authenticates requests using an API key or OIDC token. You do not need to manage provider credentials directly.
When to Use Grok 4.20 Multi-Agent
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 When
- 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
Conclusion
Grok 4.20 Multi-Agent opens up multi-agent orchestration as a first-class use case in the Grok model family. For teams building systems where agents collaborate, delegate, and coordinate, it provides a model specifically tuned for those interaction patterns. The beta status means you should plan for potential changes before committing to production use.
Frequently Asked Questions
What makes Grok 4.20 Multi-Agent different from other Grok 4.20 models?
Grok 4.20 Multi-Agent 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?
2M tokens.
Is Grok 4.20 Multi-Agent in beta?
Yes. Grok 4.20 Multi-Agent is currently in beta. Behavior, pricing, or availability may change before general availability.
What does Grok 4.20 Multi-Agent cost?
Current pricing is shown on this page. AI Gateway routes across providers, and rates may vary by provider.
How do I authenticate with Grok 4.20 Multi-Agent through Vercel AI Gateway?
Use your Vercel AI Gateway API key with
xai/grok-4.20-multi-agentas the model identifier. No separate xAI account is needed for gateway-managed access.Can I use Grok 4.20 Multi-Agent 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?
Zero Data Retention is not currently 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.