ML

Llama-4-Scout-17B-16E-Instruct

Multimodalby Meta Llama·Model page

Meta Llama's Llama-4-Scout is a multimodal MoE instruct model with 17B active parameters for image and text tasks in 12 languages.

Max output

Most tokens the model can return in a single response.

16Ktokens
Share:

Base model

meta-llama/Llama-4-Scout-17B-16E

Design Arena

Design Arena ranks models on real-world front-end and design tasks — websites, UI components, data viz, SVG and more — through head-to-head human votes, scored as an ELO rating.
CategoryELOWin rateRank
Data viz92939.3%#103
Game dev83127.4%#110
Code82626.6%#111
UI component81425.5%#105
Website78122.7%#117
Author
ML
Meta Llama
Organization · ✓
meta-llama
Details
Downloads702.9K
Likes1.3K
AccessOpen Source
Context10M tokens
Input price$0.1 /1M
Output price$0.3 /1M
Taskimage-text-to-text
Parameters109B
Licenseother
Librarytransformers
Knowledge cutoffAug 31, 2024
CreatedApr 2, 2025
UpdatedMay 22, 2025
View on Hugging Face
Benchmarks
Intelligence10.0
Coding8.2
Agentic1.1
Languages
ardeenesfrhiiditptthtlvi
Get the full context.

Sign up to read complete case studies, access detailed metrics, and unlock all use cases.

Llama-4-Scout-17B-16E-Instruct — AI Model Details | Applied