Meta Models
Meta logo

Llama 4 Scout

by Meta

Llama 4 Scout 17B Instruct (16E) is a mixture-of-experts (MoE) language model developed by Meta, activating 17 billion parameters out of a total of 109B. It supports native multimodal input (text and image) and multilingual output (text and code) across 12 supported languages. Designed for assistant-style interaction and visual reasoning, Scout uses 16 experts per forward pass and features a context length of 10 million tokens, with a training corpus of ~40 trillion tokens. Built for high efficiency and local or commercial deployment, Llama 4 Scout incorporates early fusion for seamless modality integration. It is instruction-tuned for use in multilingual chat, captioning, and image understanding tasks. Released under the Llama 4 Community License, it was last trained on data up to August 2024 and launched publicly on April 5, 2025.

Chat with Llama 4 Scout
Input Price$0.18/1M tokens
Output Price$0.66/1M tokens
Intelligence13.5
Coding6.7

Specifications

Technical details and pricing.

ProviderMeta
Context Window327,680 tokens
Release DateApr 5, 2025
ModalitiesText, Image β†’ Text
CapabilitiesVision

Benchmarks

12 benchmark scores from Artificial Analysis.

GPQA58.7%
MMLU Pro75.2%
HLE4.3%
LiveCodeBench29.9%
MATH 50084.4%
AIME 202514.0%
AIME28.3%
SciCode17.0%
LCR25.8%
IFBench39.5%
Tau215.5%
TerminalBench Hard1.5%

Composite Indices

Intelligence, Coding, Math

Standard Benchmarks

Academic and industry benchmarks

Frequently Asked Questions

What is Llama 4 Scout good for?

Use Llama 4 Scout for everyday tasks like writing, summarizing, brainstorming, and getting clear explanations.

How much does Llama 4 Scout cost?

Pricing is based on usage. Current rates are $0.18/1M tokens for input and $0.66/1M tokens for output.

Can I try Llama 4 Scout for free?

Yes. You can start a chat instantly and test the model before deciding on a plan.

Does Llama 4 Scout support images or audio?

Llama 4 Scout can understand images.

Benchmarks and pricing are sourced from Artificial Analysis where available. OpenRouter specs are used as a fallback.