# Meta: Llama 4 Scout

Provider: meta-llama  
Category: llm  
Model ID: `meta-llama/llama-4-scout`

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...

## Specs

- Context length: 327680 tokens
- Max output: 16384 tokens
- Modalities: text, image
- Released: 2025-04-05

## Pricing

- Input: $0.08 per million tokens
- Output: $0.30 per million tokens

## Providers

- **meta-llama** — ctx 327680, input $0.08/M, output $0.30/M

---
Last verified: 2026-04-23T23:46:29.618Z  
Canonical URL: https://switchy.build/models/llama-4-scout