# Llama 4 Maverick 17B Instruct

A natively multimodal Mixture of Experts (MoE) model with 17B active parameters and 128 experts. Llama 4 Maverick 17B Instruct activates a fraction of its 400B total parameters per token, which cuts inference cost compared with a dense 400B model at similar reported benchmark scores on image and text tasks.

- **Model ID:** `meta/llama-4-maverick`
- **Type:** chat
- **Providers:** bedrock
- **Context window:** 128,000
- **Maximum output tokens:** 8,192
- **Pricing:** $0.24/1M input tokens, $0.97/1M output tokens
- **Canonical page:** https://vercel.com/ai-gateway/models/llama-4-maverick

## Supported parameters

Detailed capability metadata has not been reported for this model.

## Example

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

const result = streamText({
  model: 'meta/llama-4-maverick',
  prompt: 'Why is the sky blue?'
})
```

## About

Meta released Llama 4 Maverick 17B Instruct on April 5, 2025 as one of the first two models in the Llama 4 generation. The collection is built around two architectural advances: native multimodality through early fusion, and Mixture of Experts (MoE). Llama 4 Maverick 17B Instruct is the larger and more capable of the two initial releases, with 17 billion active parameters, 128 routed experts plus one shared expert, and 400 billion total parameters. Each token activates only 17B of those 400B parameters (the shared expert plus one routed expert). This makes inference substantially more efficient than a dense 400B model while preserving the quality benefits of the larger total parameter budget.

Llama 4's native multimodality represents a different architectural approach from the adapter-based vision in Llama 3.2. Rather than adding image understanding to an existing text backbone, Llama 4 treats text and vision tokens together from the beginning in a unified backbone. This enables more coherent cross-modal reasoning.

On the LMArena leaderboard, an experimental chat version of Llama 4 Maverick 17B Instruct scored an Elo of 1417. Llama 4 Maverick 17B Instruct exceeds comparable frontier models on coding, reasoning, multilingual, long-context, and image benchmarks. It achieves results comparable to other open-weight models on reasoning and coding at less than half the active parameters.

## What to consider

For workloads that mix images and long text, Llama 4 Maverick 17B Instruct's efficiency advantage over dense models shows most at scale. Validate throughput at your expected concurrency level before you pick a provider tier. Compare $0.24 and $0.97.

## When to use

### Best For

- Production multimodal applications that pair precise image understanding with long-form text generation, product catalog processing, document analysis with mixed visual and textual content
- Creative writing, coding assistance, and multilingual applications where the MoE architecture reaches dense-model scores on published benchmarks at lower active-parameter cost
- Workloads sensitive to cost-per-output-quality where comparable-capability dense models are significantly more expensive to serve
- Long-context multimodal tasks where the combination of image and text reasoning must be maintained coherently across extended conversations
- General assistant and chat use cases where Meta designates Llama 4 Maverick 17B Instruct as the intended product workhorse

### Consider Alternatives When

- The primary requirement is handling extremely long documents: Llama 4 Scout's 10M token context window is purpose-built for that use case
- The workload is text-only and the MoE overhead of loading all experts into memory is not offset by quality gains over a dense model at similar cost
- Maximum reasoning depth is required: Llama 4 Behemoth (when available) or other frontier reasoning models may be appropriate

## Best for

- **Production multimodal applications:** Pairing image understanding with long-form text generation for product catalog processing and document analysis with mixed visual and textual content
- **Creative and coding workloads:** Multilingual applications where the MoE architecture reaches dense-model scores on published benchmarks at lower active-parameter cost
- **Cost-per-quality sensitive workloads:** Comparable-capability dense models are significantly more expensive to serve
- **Long-context multimodal tasks:** Image and text reasoning must be maintained coherently across extended conversations
- **General assistant and chat:** Meta designates Llama 4 Maverick 17B Instruct as the intended product workhorse

## Consider alternatives

- **Extreme long documents:** Llama 4 Scout's 10M token context window is purpose-built for that use case
- **Text-only workload:** The MoE overhead of loading all experts into memory is not offset by quality gains over a dense model at similar cost
- **Maximum reasoning depth:** Llama 4 Behemoth (when available) or other frontier reasoning models may be appropriate

## Frequently asked questions

### What is Mixture of Experts (MoE) and how does it work in Llama 4 Maverick 17B Instruct?

Each input token activates only a subset of the total parameters. Llama 4 Maverick 17B Instruct uses alternating dense and MoE layers. MoE layers route each token to a shared expert plus one of 128 routed experts. Only 17B of the 400B total parameters are active per token, reducing inference cost while the full parameter budget contributes to model quality.

### What does "natively multimodal" mean compared to the adapter-based vision in Llama 3.2?

Llama 3.2 added vision to an existing text backbone via cross-attention adapters, keeping language model weights frozen. Llama 4 Maverick 17B Instruct processes text and vision tokens together in a unified backbone. This enables deeper cross-modal reasoning because the model was never strictly text-only.

### What Elo score did Llama 4 Maverick 17B Instruct achieve on LMArena?

An experimental chat version of Llama 4 Maverick 17B Instruct scored an Elo of 1417 on LMArena.

### What languages does Llama 4 support?

Llama 4 supports 200 languages, including over 100 with more than 1 billion tokens each, representing 10x more multilingual coverage than Llama 3.

## Links

- [Model page](https://vercel.com/ai-gateway/models/llama-4-maverick)
- [AI Gateway documentation](https://vercel.com/docs/ai-gateway)
- [Provider model documentation](https://aws.amazon.com/bedrock/meta/)
- [Provider pricing](https://aws.amazon.com/bedrock/pricing/)