Skip to main content

Llama 4 Scout

Meta

The long-context champion. A 10-million-token window, enough to read hundreds of books at once.

Released April 5, 2025

Pricing

Input tokensN/A/M
Output tokensN/A/M

Capacity

Context window10.0M tokens

Capabilities

โœ—Reasoning & Planning
โœ—Web Search
โœ“Open Source

Best Scores

GDPval-AA v2111.0 pts
GPQA Diamond57.2%

Best For

๐Ÿ“Šanalysis
๐Ÿ“summarization
๐Ÿ”ฌresearch

Benchmark Scores

Specialized Skills

GDPval-AA v2

Real paid work from 44 different jobs, such as law, nursing, and software. Judges compare two answers side by side without knowing which model wrote them, and the winner gains rating points. A typical human expert scores 1000, so a higher number means the work was picked over a human more often.

111.0 pts

Independently measured by Artificial Analysis.

Knowledge

GPQA Diamond

PhD-level science questions written so you cannot just Google the answer. Tests whether the model can reason about hard science.

57.2%

Provider-reported; no independent run recorded yet.

Why Choose Llama 4 Scout?

Scout has the largest context window of any model: 10 million tokens. That's enough to load an entire codebase, hundreds of documents, or several books at once. Perfect for long-context retrieval and bulk processing tasks.

How It Compares

vs Llama 4 Maverick

Scout has 10x more context but is less capable on reasoning; Maverick is more general.

vs Grok 4.1 Fast

Scout is slightly longer (10M vs 2M) and more capable; Grok is cheaper.

You Might Also Like