# Grok 4.20 Multi Agent Experimental Beta API Pricing

As of October 10, 2026, Grok 4.20 Multi Agent Experimental Beta costs $1.25 per 1M input tokens and $2.50 per 1M output tokens on xAI, with cache reads at $0.20 per 1M tokens. It has a 1,000,000-token context window and returns up to 1,000,000 tokens per response.

- Page: https://models.litellm.ai/models/grok-4.20-multi-agent-experimental-beta
- Lab: xAI
- LiteLLM model name: `xai/grok-4.20-multi-agent-experimental-beta-0304`
- Context window: 1,000,000 tokens
- Max output: 1,000,000 tokens
- Added to LiteLLM: September 22, 2026
- Prices updated: October 10, 2026

## Pricing by provider

Token prices in USD per 1M tokens.

| Provider | Model name | Input | Output | Cache read | Per image | Per second | Context |
|---|---|---|---|---|---|---|---|
| xAI (snapshot 0304) | `xai/grok-4.20-multi-agent-experimental-beta-0304` | $1.25 | $2.50 | $0.20 | – | – | 1M |
| xAI (latest alias) | `xai/grok-4.20-multi-agent-experimental-beta-latest` | $1.25 | $2.50 | $0.20 | – | – | 1M |

## Cost examples

1,000 requests with 2,000 input and 500 output tokens each:

- xAI: $3.75

100 long-document requests with 100,000 input and 2,000 output tokens each:

- xAI: $13.00

## Features

Supported: Structured output, Image input, Reasoning, Prompt caching, Web search.
Not supported: Function calling.

## Use it with LiteLLM

```python
from litellm import completion

response = completion(
    model="xai/grok-4.20-multi-agent-experimental-beta-0304",
    messages=[{"role": "user", "content": "Hello!"}],
)
```

## FAQ

### How much does Grok 4.20 Multi Agent Experimental Beta cost?

Grok 4.20 Multi Agent Experimental Beta costs $1.25 per 1M input tokens and $2.50 per 1M output tokens on xAI, with cache reads at $0.20 per 1M tokens. Above 200K input tokens, input costs $2.50 and output costs $5.00 per 1M tokens.

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

Grok 4.20 Multi Agent Experimental Beta has a 1,000,000-token context window and returns up to 1,000,000 tokens per response.

### Which providers offer Grok 4.20 Multi Agent Experimental Beta?

Grok 4.20 Multi Agent Experimental Beta is available from xAI through LiteLLM, using the model name xai/grok-4.20-multi-agent-experimental-beta-0304.

### How do I call Grok 4.20 Multi Agent Experimental Beta with an OpenAI-compatible API?

Use the LiteLLM Python SDK with model="xai/grok-4.20-multi-agent-experimental-beta-0304", or add the model to the LiteLLM proxy's config.yaml and send requests to /v1/responses from any OpenAI client. LiteLLM tracks the cost of every request at the prices on this page.

### What features does Grok 4.20 Multi Agent Experimental Beta support?

Grok 4.20 Multi Agent Experimental Beta supports structured output, image input, reasoning, prompt caching and web search. It does not support function calling.

### When did LiteLLM add Grok 4.20 Multi Agent Experimental Beta?

Grok 4.20 Multi Agent Experimental Beta was added to LiteLLM's model price file on September 22, 2026.

Source: LiteLLM model price file, https://github.com/BerriAI/litellm/blob/main/model_prices_and_context_window.json