MetaMeta

Llama 4 Scout

10M-token context with MoE efficiency

Llama 4 Scout sets a record with its 10 million-token context window — enough to process entire large codebases in a single prompt. Its 16-expert MoE architecture makes it surprisingly efficient for its capability level.

Try Llama 4 Scout

Specifications

Context window10M tokens
Input price$0.050 / 1M tokens
Output price$0.170 / 1M tokens
Credits per query3 cr
Released2025-04
LicenseLlama 4 Community

Features

VisionTool useFunction callingCode

Benchmark Scores

What do these mean?

Sourced from official model cards and academic papers. Higher is better.

PhD-level science & reasoning

Mathematical olympiad problems

Chatbot Arena human preference ranking

Python coding — pass@1 rate

83%

Competition-level mathematics

General knowledge across 57 academic domains

82%

Multi-turn instruction following (out of 10)

Real-world GitHub issue resolution rate