Llama 4 Scout

MetafastOpen Source

Meta's efficient MoE model with 16 experts. 10M token context window and strong multilingual support.

Context Window

10M

tokens

Max Output

33K

tokens

Input Price

$0.15

per 1M tokens

Output Price

$0.4

per 1M tokens

Speed

198

tokens/sec

Released

Apr 2025

2025-04-05

Blended Cost

$0.28

per 1M tokens

Value Score

258.2

quality per $

Capabilities

ChatVisionCode Generation

Benchmarks

Quality Index
71
MMLU Pro
82.5
HumanEval (Coding)
81.2
MATH
68.4
Arena ELO
1195

Open Source — Licensed under Llama 4 Community

About Llama 4 Scout

Llama 4 Scout is a fast AI model by Meta, released on April 5, 2025. It supports a context window of 10000K tokens and can generate up to 33K output tokens.

At $0.15 per million input tokens and $0.4 per million output tokens, its blended cost of $0.28/1M tokens makes it one of the most affordable models available. Its value score of 258.2 reflects the balance of quality and cost.

Llama 4 Scout is available as an open-source model under the Llama 4 Community license, meaning you can self-host it for predictable costs or use it through API providers like Swfte Connect.

Using Llama 4 Scout with Swfte

Access Llama 4 Scout through Swfte Connect, our unified LLM gateway. Connect gives you a single API for 50+ models, with automatic routing, cost optimization, and fallback handling. You can also try Llama 4 Scout in our AI Playground before integrating.