Model index
Meta·Llama

Llama 4 Scout

A natively multimodal MoE with 17B active parameters, 16 experts, text-and-image input, and up to 10M context.

CurrentLanguageReasoningMultimodalOpen weights
MODEL NUMBER151
First public date
2025-04-05
Organization
Meta
Family
Llama
01

Specifications

Only published specifications are shown.

Total parameters
109B
Active parameters
17B
Architecture
MoE
Context
10,000,000 tokens
Input modalities
Text · Image
Output modalities
Text
Open weights
Yes
License
Llama 4 Community License
Status
Current
02

Overview

Release dates

— Meta released the open-weight Llama 4 Scout and Maverick multimodal MoE models. ↗

03

Specifications alongside predecessor

Compared with Llama 3.3 70B ↗

PreviousLlama 3.3 70B70B · 128K ctx
CurrentLlama 4 Scout109B · 10M ctx
04

Architecture & capabilities

Model capabilities

Long contextNative multimodal

Unclassified tags

Efficiency
05

Release history

Open weights

Meta released the open-weight Llama 4 Scout and Maverick multimodal MoE models.

View source
06

Related models

Select a card to view the related model.

07

Sources

Suggest a correction

Copy this template, add the field, proposed value, and primary source, and share it with the archive maintainer.