Meta

Llama 4 Scout

Llama 4 Scout is an open-weight model from Meta, released Apr 5, 2025. Its best result is #80 of 96 on AIME with 18.96%. It costs $0.10 per million input tokens and $0.30 per million output tokens, with a 1.3M-token context window.

Developer
Meta
Released
Apr 5, 2025
Context window
1.3M
Price in / out, per 1M tokens
$0.10 / $0.30
Weights
Open
Inputs
Text, Images
Reasoning
No
Tool use
Yes
Hugging Face downloads, 30 days
142.7K
Hugging Face likes
1.4K

Benchmarks

BenchmarkScore barAccuracyRankCost / test
LiveCodeBenchCoding38.54%132 / 143$0.18 / $0.59
MedCodeMedicine23.31%104 / 107$0.18 / $0.59
MedQAMedicine50.90%89 / 92$0.18 / $0.59
GPQA DiamondReasoning46.97%128 / 138$0.18 / $0.59
MMLU-ProReasoning69.63%123 / 138$0.18 / $0.59
AIMEReasoning18.96%80 / 96$0.18 / $0.59
MMMUReasoning58.75%85 / 93$0.18 / $0.59

Vals AI, as of Oct 24, 2026.

Compare Llama 4 Scout

Meta

Meta’s AI research publishes the Llama family, among the most downloaded open-weight models. Founded 2013 (FAIR), based in Menlo Park. ai.meta.com

Questions

Who makes Llama 4 Scout?

Llama 4 Scout is made by Meta, founded 2013 (FAIR) and based in Menlo Park.

How much does Llama 4 Scout cost?

$0.10 per million input tokens and $0.30 per million output tokens, at its provider’s list price.

What is Llama 4 Scout’s context window?

1.3M tokens (1,310,720).

Is Llama 4 Scout open source?

Llama 4 Scout has open weights, downloadable from Hugging Face (meta-llama/Llama-4-Scout-17B-16E-Instruct).

Sources

Every number on this page comes from these sources; models without a published number show none. Download the data. Cite as “Nexus Explorer, nexus.inc/models”.